Input data format (#25464)

* Add copied from statements for image processors

* Move out rescale and normalize to base image processor

* Remove rescale and normalize from vit (post rebase)

* Update docstrings and tidy up

* PR comments

* Add input_data_format as preprocess argument

* Resolve tests and tidy up

* Remove num_channels argument

* Update doc strings -> default ints not in code formatting

src/transformers/image_processing_utils.py

@@ -521,7 +521,12 @@ class BaseImageProcessor(ImageProcessingMixin):
         raise NotImplementedError("Each image processor must implement its own preprocess method")
 
     def rescale(
-        self, image: np.ndarray, scale: float, data_format: Optional[Union[str, ChannelDimension]] = None, **kwargs
+        self,
+        image: np.ndarray,
+        scale: float,
+        data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
+        **kwargs,
     ) -> np.ndarray:
         """
         Rescale an image by a scale factor. image = image * scale.
@@ -536,11 +541,16 @@ class BaseImageProcessor(ImageProcessingMixin):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
 
         Returns:
             `np.ndarray`: The rescaled image.
         """
-        return rescale(image, scale=scale, data_format=data_format, **kwargs)
+        return rescale(image, scale=scale, data_format=data_format, input_data_format=input_data_format, **kwargs)
 
     def normalize(
         self,
@@ -548,6 +558,7 @@ class BaseImageProcessor(ImageProcessingMixin):
         mean: Union[float, Iterable[float]],
         std: Union[float, Iterable[float]],
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -565,17 +576,25 @@ class BaseImageProcessor(ImageProcessingMixin):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
 
         Returns:
             `np.ndarray`: The normalized image.
         """
-        return normalize(image, mean=mean, std=std, data_format=data_format, **kwargs)
+        return normalize(
+            image, mean=mean, std=std, data_format=data_format, input_data_format=input_data_format, **kwargs
+        )
 
     def center_crop(
         self,
         image: np.ndarray,
         size: Dict[str, int],
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -588,12 +607,26 @@ class BaseImageProcessor(ImageProcessingMixin):
             size (`Dict[str, int]`):
                 Size of the output image.
             data_format (`str` or `ChannelDimension`, *optional*):
-                The channel dimension format of the image. If not provided, it will be the same as the input image.
+                The channel dimension format for the output image. If unset, the channel dimension format of the input
+                image is used. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
         """
         size = get_size_dict(size)
         if "height" not in size or "width" not in size:
             raise ValueError(f"The size dictionary must have keys 'height' and 'width'. Got {size.keys()}")
-        return center_crop(image, size=(size["height"], size["width"]), data_format=data_format, **kwargs)
+        return center_crop(
+            image,
+            size=(size["height"], size["width"]),
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
 
 VALID_SIZE_DICT_KEYS = ({"height", "width"}, {"shortest_edge"}, {"shortest_edge", "longest_edge"}, {"longest_edge"})

src/transformers/models/beit/image_processing_beit.py

@@ -27,6 +27,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -145,6 +146,7 @@ class BeitImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -159,12 +161,19 @@ class BeitImageProcessor(BaseImageProcessor):
                 Resampling filter to use when resiizing the image.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         size = get_size_dict(size, default_to_square=True, param_name="size")
         if "height" not in size or "width" not in size:
             raise ValueError(f"The `size` argument must contain `height` and `width` keys. Got {size.keys()}")
         return resize(
-            image, size=(size["height"], size["width"]), resample=resample, data_format=data_format, **kwargs
+            image,
+            size=(size["height"], size["width"]),
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
         )
 
     def reduce_label(self, label: ImageInput) -> np.ndarray:
@@ -189,21 +198,22 @@ class BeitImageProcessor(BaseImageProcessor):
         do_normalize: bool = None,
         image_mean: Optional[Union[float, List[float]]] = None,
         image_std: Optional[Union[float, List[float]]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ):
         if do_reduce_labels:
             image = self.reduce_label(image)
 
         if do_resize:
-            image = self.resize(image=image, size=size, resample=resample)
+            image = self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
 
         if do_center_crop:
-            image = self.center_crop(image=image, size=crop_size)
+            image = self.center_crop(image=image, size=crop_size, input_data_format=input_data_format)
 
         if do_rescale:
-            image = self.rescale(image=image, scale=rescale_factor)
+            image = self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
 
         if do_normalize:
-            image = self.normalize(image=image, mean=image_mean, std=image_std)
+            image = self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
 
         return image
 
@@ -221,10 +231,13 @@ class BeitImageProcessor(BaseImageProcessor):
         image_mean: Optional[Union[float, List[float]]] = None,
         image_std: Optional[Union[float, List[float]]] = None,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """Preprocesses a single image."""
         # All transformations expect numpy arrays.
         image = to_numpy_array(image)
+        if input_data_format is None:
+            input_data_format = infer_channel_dimension_format(image)
         image = self._preprocess(
             image,
             do_reduce_labels=False,
@@ -238,9 +251,10 @@ class BeitImageProcessor(BaseImageProcessor):
             do_normalize=do_normalize,
             image_mean=image_mean,
             image_std=image_std,
+            input_data_format=input_data_format,
         )
         if data_format is not None:
-            image = to_channel_dimension_format(image, data_format)
+            image = to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
         return image
 
     def _preprocess_segmentation_map(
@@ -252,6 +266,7 @@ class BeitImageProcessor(BaseImageProcessor):
         do_center_crop: bool = None,
         crop_size: Dict[str, int] = None,
         do_reduce_labels: bool = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ):
         """Preprocesses a single segmentation map."""
         # All transformations expect numpy arrays.
@@ -260,8 +275,11 @@ class BeitImageProcessor(BaseImageProcessor):
         if segmentation_map.ndim == 2:
             segmentation_map = segmentation_map[None, ...]
             added_dimension = True
+            input_data_format = ChannelDimension.FIRST
         else:
             added_dimension = False
+            if input_data_format is None:
+                input_data_format = infer_channel_dimension_format(segmentation_map, num_channels=1)
         segmentation_map = self._preprocess(
             image=segmentation_map,
             do_reduce_labels=do_reduce_labels,
@@ -272,6 +290,7 @@ class BeitImageProcessor(BaseImageProcessor):
             crop_size=crop_size,
             do_normalize=False,
             do_rescale=False,
+            input_data_format=ChannelDimension.FIRST,
         )
         # Remove extra axis if added
         if added_dimension:
@@ -301,6 +320,7 @@ class BeitImageProcessor(BaseImageProcessor):
         do_reduce_labels: Optional[bool] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -344,8 +364,15 @@ class BeitImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
                 The channel dimension format for the output image. Can be one of:
-                    - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
-                    - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size = size if size is not None else self.size
@@ -403,6 +430,7 @@ class BeitImageProcessor(BaseImageProcessor):
                 image_mean=image_mean,
                 image_std=image_std,
                 data_format=data_format,
+                input_data_format=input_data_format,
             )
             for img in images
         ]

src/transformers/models/bit/image_processing_bit.py

@@ -31,6 +31,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -125,6 +126,7 @@ class BitImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -140,12 +142,23 @@ class BitImageProcessor(BaseImageProcessor):
                 Resampling filter to use when resiizing the image.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         size = get_size_dict(size, default_to_square=False)
         if "shortest_edge" not in size:
             raise ValueError(f"The `size` parameter must contain the key `shortest_edge`. Got {size.keys()}")
-        output_size = get_resize_output_image_size(image, size=size["shortest_edge"], default_to_square=False)
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        output_size = get_resize_output_image_size(
+            image, size=size["shortest_edge"], default_to_square=False, input_data_format=input_data_format
+        )
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     def preprocess(
         self,
@@ -163,6 +176,7 @@ class BitImageProcessor(BaseImageProcessor):
         do_convert_rgb: bool = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -205,9 +219,15 @@ class BitImageProcessor(BaseImageProcessor):
                 - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
                 The channel dimension format for the output image. Can be one of:
-                - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
-                - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
-                - Unset: defaults to the channel dimension format of the input image.
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size = size if size is not None else self.size
@@ -250,19 +270,36 @@ class BitImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_center_crop:
-            images = [self.center_crop(image=image, size=crop_size) for image in images]
+            images = [
+                self.center_crop(image=image, size=crop_size, input_data_format=input_data_format) for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
 
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/blip/image_processing_blip.py

@@ -26,6 +26,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -111,6 +112,7 @@ class BlipImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -128,6 +130,13 @@ class BlipImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
 
         Returns:
             `np.ndarray`: The resized image.
@@ -136,7 +145,14 @@ class BlipImageProcessor(BaseImageProcessor):
         if "height" not in size or "width" not in size:
             raise ValueError(f"The `size` dictionary must contain the keys `height` and `width`. Got {size.keys()}")
         output_size = (size["height"], size["width"])
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     def preprocess(
         self,
@@ -152,6 +168,7 @@ class BlipImageProcessor(BaseImageProcessor):
         return_tensors: Optional[Union[str, TensorType]] = None,
         do_convert_rgb: bool = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -190,8 +207,15 @@ class BlipImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
                 The channel dimension format for the output image. Can be one of:
-                    - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
-                    - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         resample = resample if resample is not None else self.resample
@@ -229,16 +253,31 @@ class BlipImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
-
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
+
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         encoded_outputs = BatchFeature(data={"pixel_values": images}, tensor_type=return_tensors)
 

src/transformers/models/bridgetower/image_processing_bridgetower.py

@@ -50,7 +50,9 @@ def max_across_indices(values: Iterable[Any]) -> List[Any]:
 
 
 # Copied from transformers.models.vilt.image_processing_vilt.make_pixel_mask
-def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarray:
+def make_pixel_mask(
+    image: np.ndarray, output_size: Tuple[int, int], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> np.ndarray:
     """
     Make a pixel mask for the image, where 1 indicates a valid pixel and 0 indicates padding.
 
@@ -60,33 +62,40 @@ def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarr
         output_size (`Tuple[int, int]`):
             Output size of the mask.
     """
-    input_height, input_width = get_image_size(image)
+    input_height, input_width = get_image_size(image, channel_dim=input_data_format)
     mask = np.zeros(output_size, dtype=np.int64)
     mask[:input_height, :input_width] = 1
     return mask
 
 
 # Copied from transformers.models.vilt.image_processing_vilt.get_max_height_width
-def get_max_height_width(images: List[np.ndarray]) -> List[int]:
+def get_max_height_width(
+    images: List[np.ndarray], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> List[int]:
     """
     Get the maximum height and width across all images in a batch.
     """
-    input_channel_dimension = infer_channel_dimension_format(images[0])
+    if input_data_format is None:
+        input_data_format = infer_channel_dimension_format(images[0])
 
-    if input_channel_dimension == ChannelDimension.FIRST:
+    if input_data_format == ChannelDimension.FIRST:
         _, max_height, max_width = max_across_indices([img.shape for img in images])
-    elif input_channel_dimension == ChannelDimension.LAST:
+    elif input_data_format == ChannelDimension.LAST:
         max_height, max_width, _ = max_across_indices([img.shape for img in images])
     else:
-        raise ValueError(f"Invalid channel dimension format: {input_channel_dimension}")
+        raise ValueError(f"Invalid channel dimension format: {input_data_format}")
     return (max_height, max_width)
 
 
 # Copied from transformers.models.vilt.image_processing_vilt.get_resize_output_image_size
 def get_resize_output_image_size(
-    input_image: np.ndarray, shorter: int = 800, longer: int = 1333, size_divisor: int = 32
+    input_image: np.ndarray,
+    shorter: int = 800,
+    longer: int = 1333,
+    size_divisor: int = 32,
+    input_data_format: Optional[Union[str, ChannelDimension]] = None,
 ) -> Tuple[int, int]:
-    input_height, input_width = get_image_size(input_image)
+    input_height, input_width = get_image_size(input_image, input_data_format)
     min_size, max_size = shorter, longer
 
     scale = min_size / min(input_height, input_width)
@@ -122,7 +131,7 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
             Resize the shorter side of the input to `size["shortest_edge"]`. The longer side will be limited to under
             `int((1333 / 800) * size["shortest_edge"])` while preserving the aspect ratio. Only has an effect if
             `do_resize` is set to `True`. Can be overridden by the `size` parameter in the `preprocess` method.
-        size_divisor (`int`, *optional*, defaults to `32`):
+        size_divisor (`int`, *optional*, defaults to 32):
             The size by which to make sure both the height and width can be divided. Only has an effect if `do_resize`
             is set to `True`. Can be overridden by the `size_divisor` parameter in the `preprocess` method.
         resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BICUBIC`):
@@ -197,6 +206,7 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
         size_divisor: int = 32,
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -217,20 +227,32 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
                 Resampling filter to use when resiizing the image.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         size = get_size_dict(size, default_to_square=False)
         if "shortest_edge" not in size:
             raise ValueError(f"The `size` dictionary must contain the key `shortest_edge`. Got {size.keys()}")
         shorter = size["shortest_edge"]
         longer = int(1333 / 800 * shorter)
-        output_size = get_resize_output_image_size(image, shorter=shorter, longer=longer, size_divisor=size_divisor)
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        output_size = get_resize_output_image_size(
+            image, shorter=shorter, longer=longer, size_divisor=size_divisor, input_data_format=input_data_format
+        )
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     def center_crop(
         self,
         image: np.ndarray,
         size: Dict[str, int],
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -244,9 +266,18 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
                 Size of the output image in the form `{"height": h, "width": w}`.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred from the input
+                image.
         """
         output_size = size["shortest_edge"]
-        return center_crop(image, size=(output_size, output_size), data_format=data_format, **kwargs)
+        return center_crop(
+            image,
+            size=(output_size, output_size),
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor._pad_image
     def _pad_image(
@@ -255,18 +286,24 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
         output_size: Tuple[int, int],
         constant_values: Union[float, Iterable[float]] = 0,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Pad an image with zeros to the given size.
         """
-        input_height, input_width = get_image_size(image)
+        input_height, input_width = get_image_size(image, channel_dim=input_data_format)
         output_height, output_width = output_size
 
         pad_bottom = output_height - input_height
         pad_right = output_width - input_width
         padding = ((0, pad_bottom), (0, pad_right))
         padded_image = pad(
-            image, padding, mode=PaddingMode.CONSTANT, constant_values=constant_values, data_format=data_format
+            image,
+            padding,
+            mode=PaddingMode.CONSTANT,
+            constant_values=constant_values,
+            data_format=data_format,
+            input_data_format=input_data_format,
         )
         return padded_image
 
@@ -278,6 +315,7 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
         return_pixel_mask: bool = True,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> BatchFeature:
         """
         Pads a batch of images to the bottom and right of the image with zeros to the size of largest height and width
@@ -299,17 +337,28 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
-        pad_size = get_max_height_width(images)
+        pad_size = get_max_height_width(images, input_data_format=input_data_format)
 
         padded_images = [
-            self._pad_image(image, pad_size, constant_values=constant_values, data_format=data_format)
+            self._pad_image(
+                image,
+                pad_size,
+                constant_values=constant_values,
+                data_format=data_format,
+                input_data_format=input_data_format,
+            )
             for image in images
         ]
         data = {"pixel_values": padded_images}
 
         if return_pixel_mask:
-            masks = [make_pixel_mask(image=image, output_size=pad_size) for image in images]
+            masks = [
+                make_pixel_mask(image=image, output_size=pad_size, input_data_format=input_data_format)
+                for image in images
+            ]
             data["pixel_mask"] = masks
 
         return BatchFeature(data=data, tensor_type=return_tensors)
@@ -330,6 +379,7 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
         do_center_crop: Optional[bool] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -374,8 +424,15 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
                 The channel dimension format for the output image. Can be one of:
-                    - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
-                    - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size_divisor = size_divisor if size_divisor is not None else self.size_divisor
@@ -414,22 +471,41 @@ class BridgeTowerImageProcessor(BaseImageProcessor):
 
         if do_resize:
             images = [
-                self.resize(image=image, size=size, size_divisor=size_divisor, resample=resample) for image in images
+                self.resize(
+                    image=image,
+                    size=size,
+                    size_divisor=size_divisor,
+                    resample=resample,
+                    input_data_format=input_data_format,
+                )
+                for image in images
             ]
 
         if do_center_crop:
-            images = [self.center_crop(image=image, size=size) for image in images]
+            images = [
+                self.center_crop(image=image, size=size, input_data_format=input_data_format) for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
 
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         if do_pad:
-            encoded_outputs = self.pad(images, return_pixel_mask=True, return_tensors=return_tensors)
+            encoded_outputs = self.pad(
+                images, return_pixel_mask=True, return_tensors=return_tensors, input_data_format=data_format
+            )
         else:
             encoded_outputs = BatchFeature(data={"pixel_values": images}, tensor_type=return_tensors)
 

src/transformers/models/chinese_clip/image_processing_chinese_clip.py

@@ -31,6 +31,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -124,6 +125,7 @@ class ChineseCLIPImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -139,12 +141,22 @@ class ChineseCLIPImageProcessor(BaseImageProcessor):
                 Resampling filter to use when resiizing the image.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred from the input
+                image.
         """
         size = get_size_dict(size, default_to_square=False)
         output_size = get_resize_output_image_size(
-            image, size=(size["height"], size["width"]), default_to_square=False
+            image, size=(size["height"], size["width"]), default_to_square=False, input_data_format=input_data_format
+        )
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
         )
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
 
     def preprocess(
         self,
@@ -162,6 +174,7 @@ class ChineseCLIPImageProcessor(BaseImageProcessor):
         do_convert_rgb: bool = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -204,9 +217,15 @@ class ChineseCLIPImageProcessor(BaseImageProcessor):
                 - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
                 The channel dimension format for the output image. Can be one of:
-                - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
-                - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
-                - Unset: defaults to the channel dimension format of the input image.
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size = size if size is not None else self.size
@@ -249,19 +268,36 @@ class ChineseCLIPImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_center_crop:
-            images = [self.center_crop(image=image, size=crop_size) for image in images]
+            images = [
+                self.center_crop(image=image, size=crop_size, input_data_format=input_data_format) for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
 
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/clip/image_processing_clip.py

@@ -31,6 +31,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -124,6 +125,7 @@ class CLIPImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -139,12 +141,23 @@ class CLIPImageProcessor(BaseImageProcessor):
                 Resampling filter to use when resiizing the image.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         size = get_size_dict(size, default_to_square=False)
         if "shortest_edge" not in size:
             raise ValueError(f"The `size` parameter must contain the key `shortest_edge`. Got {size.keys()}")
-        output_size = get_resize_output_image_size(image, size=size["shortest_edge"], default_to_square=False)
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        output_size = get_resize_output_image_size(
+            image, size=size["shortest_edge"], default_to_square=False, input_data_format=input_data_format
+        )
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     def preprocess(
         self,
@@ -162,6 +175,7 @@ class CLIPImageProcessor(BaseImageProcessor):
         do_convert_rgb: bool = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -204,9 +218,15 @@ class CLIPImageProcessor(BaseImageProcessor):
                 - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
                 The channel dimension format for the output image. Can be one of:
-                - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
-                - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
-                - Unset: defaults to the channel dimension format of the input image.
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size = size if size is not None else self.size
@@ -249,19 +269,36 @@ class CLIPImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_center_crop:
-            images = [self.center_crop(image=image, size=crop_size) for image in images]
+            images = [
+                self.center_crop(image=image, size=crop_size, input_data_format=input_data_format) for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
 
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/conditional_detr/image_processing_conditional_detr.py

@@ -124,7 +124,10 @@ def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple[int, in
 
 # Copied from transformers.models.detr.image_processing_detr.get_resize_output_image_size
 def get_resize_output_image_size(
-    input_image: np.ndarray, size: Union[int, Tuple[int, int], List[int]], max_size: Optional[int] = None
+    input_image: np.ndarray,
+    size: Union[int, Tuple[int, int], List[int]],
+    max_size: Optional[int] = None,
+    input_data_format: Optional[Union[str, ChannelDimension]] = None,
 ) -> Tuple[int, int]:
     """
     Computes the output image size given the input image size and the desired output size. If the desired output size
@@ -138,8 +141,10 @@ def get_resize_output_image_size(
             The desired output size.
         max_size (`int`, *optional*):
             The maximum allowed output size.
+        input_data_format (`ChannelDimension` or `str`, *optional*):
+            The channel dimension format of the input image. If not provided, it will be inferred from the input image.
     """
-    image_size = get_image_size(input_image)
+    image_size = get_image_size(input_image, input_data_format)
     if isinstance(size, (list, tuple)):
         return size
 
@@ -209,23 +214,28 @@ def max_across_indices(values: Iterable[Any]) -> List[Any]:
 
 
 # Copied from transformers.models.detr.image_processing_detr.get_max_height_width
-def get_max_height_width(images: List[np.ndarray]) -> List[int]:
+def get_max_height_width(
+    images: List[np.ndarray], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> List[int]:
     """
     Get the maximum height and width across all images in a batch.
     """
-    input_channel_dimension = infer_channel_dimension_format(images[0])
+    if input_data_format is None:
+        input_data_format = infer_channel_dimension_format(images[0])
 
-    if input_channel_dimension == ChannelDimension.FIRST:
+    if input_data_format == ChannelDimension.FIRST:
         _, max_height, max_width = max_across_indices([img.shape for img in images])
-    elif input_channel_dimension == ChannelDimension.LAST:
+    elif input_data_format == ChannelDimension.LAST:
         max_height, max_width, _ = max_across_indices([img.shape for img in images])
     else:
-        raise ValueError(f"Invalid channel dimension format: {input_channel_dimension}")
+        raise ValueError(f"Invalid channel dimension format: {input_data_format}")
     return (max_height, max_width)
 
 
 # Copied from transformers.models.detr.image_processing_detr.make_pixel_mask
-def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarray:
+def make_pixel_mask(
+    image: np.ndarray, output_size: Tuple[int, int], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> np.ndarray:
     """
     Make a pixel mask for the image, where 1 indicates a valid pixel and 0 indicates padding.
 
@@ -235,7 +245,7 @@ def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarr
         output_size (`Tuple[int, int]`):
             Output size of the mask.
     """
-    input_height, input_width = get_image_size(image)
+    input_height, input_width = get_image_size(image, channel_dim=input_data_format)
     mask = np.zeros(output_size, dtype=np.int64)
     mask[:input_height, :input_width] = 1
     return mask
@@ -277,11 +287,16 @@ def convert_coco_poly_to_mask(segmentations, height: int, width: int) -> np.ndar
 
 
 # Copied from transformers.models.detr.image_processing_detr.prepare_coco_detection_annotation with DETR->ConditionalDetr
-def prepare_coco_detection_annotation(image, target, return_segmentation_masks: bool = False):
+def prepare_coco_detection_annotation(
+    image,
+    target,
+    return_segmentation_masks: bool = False,
+    input_data_format: Optional[Union[ChannelDimension, str]] = None,
+):
     """
     Convert the target in COCO format into the format expected by ConditionalDetr.
     """
-    image_height, image_width = get_image_size(image)
+    image_height, image_width = get_image_size(image, channel_dim=input_data_format)
 
     image_id = target["image_id"]
     image_id = np.asarray([image_id], dtype=np.int64)
@@ -366,12 +381,16 @@ def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
 
 # Copied from transformers.models.detr.image_processing_detr.prepare_coco_panoptic_annotation with DETR->ConditionalDetr
 def prepare_coco_panoptic_annotation(
-    image: np.ndarray, target: Dict, masks_path: Union[str, pathlib.Path], return_masks: bool = True
+    image: np.ndarray,
+    target: Dict,
+    masks_path: Union[str, pathlib.Path],
+    return_masks: bool = True,
+    input_data_format: Union[ChannelDimension, str] = None,
 ) -> Dict:
     """
     Prepare a coco panoptic annotation for ConditionalDetr.
     """
-    image_height, image_width = get_image_size(image)
+    image_height, image_width = get_image_size(image, channel_dim=input_data_format)
     annotation_path = pathlib.Path(masks_path) / target["file_name"]
 
     new_target = {}
@@ -842,6 +861,7 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
         format: Optional[AnnotionFormat] = None,
         return_segmentation_masks: bool = None,
         masks_path: Optional[Union[str, pathlib.Path]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> Dict:
         """
         Prepare an annotation for feeding into ConditionalDetr model.
@@ -850,11 +870,17 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
 
         if format == AnnotionFormat.COCO_DETECTION:
             return_segmentation_masks = False if return_segmentation_masks is None else return_segmentation_masks
-            target = prepare_coco_detection_annotation(image, target, return_segmentation_masks)
+            target = prepare_coco_detection_annotation(
+                image, target, return_segmentation_masks, input_data_format=input_data_format
+            )
         elif format == AnnotionFormat.COCO_PANOPTIC:
             return_segmentation_masks = True if return_segmentation_masks is None else return_segmentation_masks
             target = prepare_coco_panoptic_annotation(
-                image, target, masks_path=masks_path, return_masks=return_segmentation_masks
+                image,
+                target,
+                masks_path=masks_path,
+                return_masks=return_segmentation_masks,
+                input_data_format=input_data_format,
             )
         else:
             raise ValueError(f"Format {format} is not supported.")
@@ -892,11 +918,26 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BILINEAR,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
         Resize the image to the given size. Size can be `min_size` (scalar) or `(height, width)` tuple. If size is an
         int, smaller edge of the image will be matched to this number.
+
+        Args:
+            image (`np.ndarray`):
+                Image to resize.
+            size (`Dict[str, int]`):
+                Dictionary containing the size to resize to. Can contain the keys `shortest_edge` and `longest_edge` or
+                `height` and `width`.
+            resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`):
+                Resampling filter to use if resizing the image.
+            data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the output image. If unset, the channel dimension format of the input
+                image is used.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         if "max_size" in kwargs:
             logger.warning_once(
@@ -908,7 +949,9 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
             max_size = None
         size = get_size_dict(size, max_size=max_size, default_to_square=False)
         if "shortest_edge" in size and "longest_edge" in size:
-            size = get_resize_output_image_size(image, size["shortest_edge"], size["longest_edge"])
+            size = get_resize_output_image_size(
+                image, size["shortest_edge"], size["longest_edge"], input_data_format=input_data_format
+            )
         elif "height" in size and "width" in size:
             size = (size["height"], size["width"])
         else:
@@ -916,7 +959,9 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
                 "Size must contain 'height' and 'width' keys or 'shortest_edge' and 'longest_edge' keys. Got"
                 f" {size.keys()}."
             )
-        image = resize(image, size=size, resample=resample, data_format=data_format)
+        image = resize(
+            image, size=size, resample=resample, data_format=data_format, input_data_format=input_data_format, **kwargs
+        )
         return image
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.resize_annotation
@@ -935,7 +980,11 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.rescale
     def rescale(
-        self, image: np.ndarray, rescale_factor: float, data_format: Optional[Union[str, ChannelDimension]] = None
+        self,
+        image: np.ndarray,
+        rescale_factor: float,
+        data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Rescale the image by the given factor. image = image * rescale_factor.
@@ -950,8 +999,13 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the input image. If unset, is inferred from the input image. Can be
+                one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
         """
-        return rescale(image, rescale_factor, data_format=data_format)
+        return rescale(image, rescale_factor, data_format=data_format, input_data_format=input_data_format)
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.normalize_annotation
     def normalize_annotation(self, annotation: Dict, image_size: Tuple[int, int]) -> Dict:
@@ -968,18 +1022,24 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
         output_size: Tuple[int, int],
         constant_values: Union[float, Iterable[float]] = 0,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Pad an image with zeros to the given size.
         """
-        input_height, input_width = get_image_size(image)
+        input_height, input_width = get_image_size(image, channel_dim=input_data_format)
         output_height, output_width = output_size
 
         pad_bottom = output_height - input_height
         pad_right = output_width - input_width
         padding = ((0, pad_bottom), (0, pad_right))
         padded_image = pad(
-            image, padding, mode=PaddingMode.CONSTANT, constant_values=constant_values, data_format=data_format
+            image,
+            padding,
+            mode=PaddingMode.CONSTANT,
+            constant_values=constant_values,
+            data_format=data_format,
+            input_data_format=input_data_format,
         )
         return padded_image
 
@@ -991,6 +1051,7 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
         return_pixel_mask: bool = True,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> BatchFeature:
         """
         Pads a batch of images to the bottom and right of the image with zeros to the size of largest height and width
@@ -1012,17 +1073,28 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
-        pad_size = get_max_height_width(images)
+        pad_size = get_max_height_width(images, input_data_format=input_data_format)
 
         padded_images = [
-            self._pad_image(image, pad_size, constant_values=constant_values, data_format=data_format)
+            self._pad_image(
+                image,
+                pad_size,
+                constant_values=constant_values,
+                data_format=data_format,
+                input_data_format=input_data_format,
+            )
             for image in images
         ]
         data = {"pixel_values": padded_images}
 
         if return_pixel_mask:
-            masks = [make_pixel_mask(image=image, output_size=pad_size) for image in images]
+            masks = [
+                make_pixel_mask(image=image, output_size=pad_size, input_data_format=input_data_format)
+                for image in images
+            ]
             data["pixel_mask"] = masks
 
         return BatchFeature(data=data, tensor_type=return_tensors)
@@ -1046,6 +1118,7 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
         format: Optional[Union[str, AnnotionFormat]] = None,
         return_tensors: Optional[Union[TensorType, str]] = None,
         data_format: Union[str, ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> BatchFeature:
         """
@@ -1091,8 +1164,17 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
                 Format of the annotations.
             return_tensors (`str` or `TensorType`, *optional*, defaults to self.return_tensors):
                 Type of tensors to return. If `None`, will return the list of images.
-            data_format (`str` or `ChannelDimension`, *optional*, defaults to self.data_format):
-                The channel dimension format of the image. If not provided, it will be the same as the input image.
+            data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
+                The channel dimension format for the output image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         if "pad_and_return_pixel_mask" in kwargs:
             logger.warning_once(
@@ -1177,13 +1259,22 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         # prepare (COCO annotations as a list of Dict -> DETR target as a single Dict per image)
         if annotations is not None:
             prepared_images = []
             prepared_annotations = []
             for image, target in zip(images, annotations):
                 target = self.prepare_annotation(
-                    image, target, format, return_segmentation_masks=return_segmentation_masks, masks_path=masks_path
+                    image,
+                    target,
+                    format,
+                    return_segmentation_masks=return_segmentation_masks,
+                    masks_path=masks_path,
+                    input_data_format=input_data_format,
                 )
                 prepared_images.append(image)
                 prepared_annotations.append(target)
@@ -1196,33 +1287,49 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
             if annotations is not None:
                 resized_images, resized_annotations = [], []
                 for image, target in zip(images, annotations):
-                    orig_size = get_image_size(image)
-                    resized_image = self.resize(image, size=size, max_size=max_size, resample=resample)
-                    resized_annotation = self.resize_annotation(target, orig_size, get_image_size(resized_image))
+                    orig_size = get_image_size(image, input_data_format)
+                    resized_image = self.resize(
+                        image, size=size, max_size=max_size, resample=resample, input_data_format=input_data_format
+                    )
+                    resized_annotation = self.resize_annotation(
+                        target, orig_size, get_image_size(resized_image, input_data_format)
+                    )
                     resized_images.append(resized_image)
                     resized_annotations.append(resized_annotation)
                 images = resized_images
                 annotations = resized_annotations
                 del resized_images, resized_annotations
             else:
-                images = [self.resize(image, size=size, resample=resample) for image in images]
+                images = [
+                    self.resize(image, size=size, resample=resample, input_data_format=input_data_format)
+                    for image in images
+                ]
 
         if do_rescale:
-            images = [self.rescale(image, rescale_factor) for image in images]
+            images = [self.rescale(image, rescale_factor, input_data_format=input_data_format) for image in images]
 
         if do_normalize:
-            images = [self.normalize(image, image_mean, image_std) for image in images]
+            images = [
+                self.normalize(image, image_mean, image_std, input_data_format=input_data_format) for image in images
+            ]
             if annotations is not None:
                 annotations = [
-                    self.normalize_annotation(annotation, get_image_size(image))
+                    self.normalize_annotation(
+                        annotation, get_image_size(image, input_data_format), input_data_format=input_data_format
+                    )
                     for annotation, image in zip(annotations, images)
                 ]
 
         if do_pad:
             # Pads images and returns their mask: {'pixel_values': ..., 'pixel_mask': ...}
-            data = self.pad(images, return_pixel_mask=True, data_format=data_format)
+            data = self.pad(
+                images, return_pixel_mask=True, data_format=data_format, input_data_format=input_data_format
+            )
         else:
-            images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
+                for image in images
+            ]
             data = {"pixel_values": images}
 
         encoded_inputs = BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/convnext/image_processing_convnext.py

@@ -31,6 +31,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -118,6 +119,7 @@ class ConvNextImageProcessor(BaseImageProcessor):
         crop_pct: float,
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -137,6 +139,9 @@ class ConvNextImageProcessor(BaseImageProcessor):
                 Resampling filter to use when resizing the image.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred from the input
+                image.
         """
         size = get_size_dict(size, default_to_square=False)
         if "shortest_edge" not in size:
@@ -146,14 +151,34 @@ class ConvNextImageProcessor(BaseImageProcessor):
         if shortest_edge < 384:
             # maintain same ratio, resizing shortest edge to shortest_edge/crop_pct
             resize_shortest_edge = int(shortest_edge / crop_pct)
-            resize_size = get_resize_output_image_size(image, size=resize_shortest_edge, default_to_square=False)
-            image = resize(image=image, size=resize_size, resample=resample, data_format=data_format, **kwargs)
+            resize_size = get_resize_output_image_size(
+                image, size=resize_shortest_edge, default_to_square=False, input_data_format=input_data_format
+            )
+            image = resize(
+                image=image,
+                size=resize_size,
+                resample=resample,
+                data_format=data_format,
+                input_data_format=input_data_format,
+                **kwargs,
+            )
             # then crop to (shortest_edge, shortest_edge)
-            return center_crop(image=image, size=(shortest_edge, shortest_edge), data_format=data_format, **kwargs)
+            return center_crop(
+                image=image,
+                size=(shortest_edge, shortest_edge),
+                data_format=data_format,
+                input_data_format=input_data_format,
+                **kwargs,
+            )
         else:
             # warping (no cropping) when evaluated at 384 or larger
             return resize(
-                image, size=(shortest_edge, shortest_edge), resample=resample, data_format=data_format, **kwargs
+                image,
+                size=(shortest_edge, shortest_edge),
+                resample=resample,
+                data_format=data_format,
+                input_data_format=input_data_format,
+                **kwargs,
             )
 
     def preprocess(
@@ -170,6 +195,7 @@ class ConvNextImageProcessor(BaseImageProcessor):
         image_std: Optional[Union[float, List[float]]] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -209,8 +235,15 @@ class ConvNextImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
                 The channel dimension format for the output image. Can be one of:
-                    - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
-                    - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         crop_pct = crop_pct if crop_pct is not None else self.crop_pct
@@ -247,16 +280,33 @@ class ConvNextImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, crop_pct=crop_pct, resample=resample) for image in images]
+            images = [
+                self.resize(
+                    image=image, size=size, crop_pct=crop_pct, resample=resample, input_data_format=input_data_format
+                )
+                for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
 
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/deformable_detr/image_processing_deformable_detr.py

@@ -123,7 +123,10 @@ def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple[int, in
 
 # Copied from transformers.models.detr.image_processing_detr.get_resize_output_image_size
 def get_resize_output_image_size(
-    input_image: np.ndarray, size: Union[int, Tuple[int, int], List[int]], max_size: Optional[int] = None
+    input_image: np.ndarray,
+    size: Union[int, Tuple[int, int], List[int]],
+    max_size: Optional[int] = None,
+    input_data_format: Optional[Union[str, ChannelDimension]] = None,
 ) -> Tuple[int, int]:
     """
     Computes the output image size given the input image size and the desired output size. If the desired output size
@@ -137,8 +140,10 @@ def get_resize_output_image_size(
             The desired output size.
         max_size (`int`, *optional*):
             The maximum allowed output size.
+        input_data_format (`ChannelDimension` or `str`, *optional*):
+            The channel dimension format of the input image. If not provided, it will be inferred from the input image.
     """
-    image_size = get_image_size(input_image)
+    image_size = get_image_size(input_image, input_data_format)
     if isinstance(size, (list, tuple)):
         return size
 
@@ -208,23 +213,28 @@ def max_across_indices(values: Iterable[Any]) -> List[Any]:
 
 
 # Copied from transformers.models.detr.image_processing_detr.get_max_height_width
-def get_max_height_width(images: List[np.ndarray]) -> List[int]:
+def get_max_height_width(
+    images: List[np.ndarray], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> List[int]:
     """
     Get the maximum height and width across all images in a batch.
     """
-    input_channel_dimension = infer_channel_dimension_format(images[0])
+    if input_data_format is None:
+        input_data_format = infer_channel_dimension_format(images[0])
 
-    if input_channel_dimension == ChannelDimension.FIRST:
+    if input_data_format == ChannelDimension.FIRST:
         _, max_height, max_width = max_across_indices([img.shape for img in images])
-    elif input_channel_dimension == ChannelDimension.LAST:
+    elif input_data_format == ChannelDimension.LAST:
         max_height, max_width, _ = max_across_indices([img.shape for img in images])
     else:
-        raise ValueError(f"Invalid channel dimension format: {input_channel_dimension}")
+        raise ValueError(f"Invalid channel dimension format: {input_data_format}")
     return (max_height, max_width)
 
 
 # Copied from transformers.models.detr.image_processing_detr.make_pixel_mask
-def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarray:
+def make_pixel_mask(
+    image: np.ndarray, output_size: Tuple[int, int], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> np.ndarray:
     """
     Make a pixel mask for the image, where 1 indicates a valid pixel and 0 indicates padding.
 
@@ -234,7 +244,7 @@ def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarr
         output_size (`Tuple[int, int]`):
             Output size of the mask.
     """
-    input_height, input_width = get_image_size(image)
+    input_height, input_width = get_image_size(image, channel_dim=input_data_format)
     mask = np.zeros(output_size, dtype=np.int64)
     mask[:input_height, :input_width] = 1
     return mask
@@ -276,11 +286,16 @@ def convert_coco_poly_to_mask(segmentations, height: int, width: int) -> np.ndar
 
 
 # Copied from transformers.models.detr.image_processing_detr.prepare_coco_detection_annotation with DETR->DeformableDetr
-def prepare_coco_detection_annotation(image, target, return_segmentation_masks: bool = False):
+def prepare_coco_detection_annotation(
+    image,
+    target,
+    return_segmentation_masks: bool = False,
+    input_data_format: Optional[Union[ChannelDimension, str]] = None,
+):
     """
     Convert the target in COCO format into the format expected by DeformableDetr.
     """
-    image_height, image_width = get_image_size(image)
+    image_height, image_width = get_image_size(image, channel_dim=input_data_format)
 
     image_id = target["image_id"]
     image_id = np.asarray([image_id], dtype=np.int64)
@@ -365,12 +380,16 @@ def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
 
 # Copied from transformers.models.detr.image_processing_detr.prepare_coco_panoptic_annotation with DETR->DeformableDetr
 def prepare_coco_panoptic_annotation(
-    image: np.ndarray, target: Dict, masks_path: Union[str, pathlib.Path], return_masks: bool = True
+    image: np.ndarray,
+    target: Dict,
+    masks_path: Union[str, pathlib.Path],
+    return_masks: bool = True,
+    input_data_format: Union[ChannelDimension, str] = None,
 ) -> Dict:
     """
     Prepare a coco panoptic annotation for DeformableDetr.
     """
-    image_height, image_width = get_image_size(image)
+    image_height, image_width = get_image_size(image, channel_dim=input_data_format)
     annotation_path = pathlib.Path(masks_path) / target["file_name"]
 
     new_target = {}
@@ -840,6 +859,7 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
         format: Optional[AnnotionFormat] = None,
         return_segmentation_masks: bool = None,
         masks_path: Optional[Union[str, pathlib.Path]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> Dict:
         """
         Prepare an annotation for feeding into DeformableDetr model.
@@ -848,11 +868,17 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
 
         if format == AnnotionFormat.COCO_DETECTION:
             return_segmentation_masks = False if return_segmentation_masks is None else return_segmentation_masks
-            target = prepare_coco_detection_annotation(image, target, return_segmentation_masks)
+            target = prepare_coco_detection_annotation(
+                image, target, return_segmentation_masks, input_data_format=input_data_format
+            )
         elif format == AnnotionFormat.COCO_PANOPTIC:
             return_segmentation_masks = True if return_segmentation_masks is None else return_segmentation_masks
             target = prepare_coco_panoptic_annotation(
-                image, target, masks_path=masks_path, return_masks=return_segmentation_masks
+                image,
+                target,
+                masks_path=masks_path,
+                return_masks=return_segmentation_masks,
+                input_data_format=input_data_format,
             )
         else:
             raise ValueError(f"Format {format} is not supported.")
@@ -890,11 +916,26 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BILINEAR,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
         Resize the image to the given size. Size can be `min_size` (scalar) or `(height, width)` tuple. If size is an
         int, smaller edge of the image will be matched to this number.
+
+        Args:
+            image (`np.ndarray`):
+                Image to resize.
+            size (`Dict[str, int]`):
+                Dictionary containing the size to resize to. Can contain the keys `shortest_edge` and `longest_edge` or
+                `height` and `width`.
+            resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`):
+                Resampling filter to use if resizing the image.
+            data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the output image. If unset, the channel dimension format of the input
+                image is used.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         if "max_size" in kwargs:
             logger.warning_once(
@@ -906,7 +947,9 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
             max_size = None
         size = get_size_dict(size, max_size=max_size, default_to_square=False)
         if "shortest_edge" in size and "longest_edge" in size:
-            size = get_resize_output_image_size(image, size["shortest_edge"], size["longest_edge"])
+            size = get_resize_output_image_size(
+                image, size["shortest_edge"], size["longest_edge"], input_data_format=input_data_format
+            )
         elif "height" in size and "width" in size:
             size = (size["height"], size["width"])
         else:
@@ -914,7 +957,9 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
                 "Size must contain 'height' and 'width' keys or 'shortest_edge' and 'longest_edge' keys. Got"
                 f" {size.keys()}."
             )
-        image = resize(image, size=size, resample=resample, data_format=data_format)
+        image = resize(
+            image, size=size, resample=resample, data_format=data_format, input_data_format=input_data_format, **kwargs
+        )
         return image
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.resize_annotation
@@ -933,7 +978,11 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.rescale
     def rescale(
-        self, image: np.ndarray, rescale_factor: float, data_format: Optional[Union[str, ChannelDimension]] = None
+        self,
+        image: np.ndarray,
+        rescale_factor: float,
+        data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Rescale the image by the given factor. image = image * rescale_factor.
@@ -948,8 +997,13 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the input image. If unset, is inferred from the input image. Can be
+                one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
         """
-        return rescale(image, rescale_factor, data_format=data_format)
+        return rescale(image, rescale_factor, data_format=data_format, input_data_format=input_data_format)
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.normalize_annotation
     def normalize_annotation(self, annotation: Dict, image_size: Tuple[int, int]) -> Dict:
@@ -966,18 +1020,24 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
         output_size: Tuple[int, int],
         constant_values: Union[float, Iterable[float]] = 0,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Pad an image with zeros to the given size.
         """
-        input_height, input_width = get_image_size(image)
+        input_height, input_width = get_image_size(image, channel_dim=input_data_format)
         output_height, output_width = output_size
 
         pad_bottom = output_height - input_height
         pad_right = output_width - input_width
         padding = ((0, pad_bottom), (0, pad_right))
         padded_image = pad(
-            image, padding, mode=PaddingMode.CONSTANT, constant_values=constant_values, data_format=data_format
+            image,
+            padding,
+            mode=PaddingMode.CONSTANT,
+            constant_values=constant_values,
+            data_format=data_format,
+            input_data_format=input_data_format,
         )
         return padded_image
 
@@ -989,6 +1049,7 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
         return_pixel_mask: bool = True,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> BatchFeature:
         """
         Pads a batch of images to the bottom and right of the image with zeros to the size of largest height and width
@@ -1010,17 +1071,28 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
-        pad_size = get_max_height_width(images)
+        pad_size = get_max_height_width(images, input_data_format=input_data_format)
 
         padded_images = [
-            self._pad_image(image, pad_size, constant_values=constant_values, data_format=data_format)
+            self._pad_image(
+                image,
+                pad_size,
+                constant_values=constant_values,
+                data_format=data_format,
+                input_data_format=input_data_format,
+            )
             for image in images
         ]
         data = {"pixel_values": padded_images}
 
         if return_pixel_mask:
-            masks = [make_pixel_mask(image=image, output_size=pad_size) for image in images]
+            masks = [
+                make_pixel_mask(image=image, output_size=pad_size, input_data_format=input_data_format)
+                for image in images
+            ]
             data["pixel_mask"] = masks
 
         return BatchFeature(data=data, tensor_type=return_tensors)
@@ -1044,6 +1116,7 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
         format: Optional[Union[str, AnnotionFormat]] = None,
         return_tensors: Optional[Union[TensorType, str]] = None,
         data_format: Union[str, ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> BatchFeature:
         """
@@ -1089,8 +1162,17 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
                 Format of the annotations.
             return_tensors (`str` or `TensorType`, *optional*, defaults to self.return_tensors):
                 Type of tensors to return. If `None`, will return the list of images.
-            data_format (`str` or `ChannelDimension`, *optional*, defaults to self.data_format):
-                The channel dimension format of the image. If not provided, it will be the same as the input image.
+            data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
+                The channel dimension format for the output image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         if "pad_and_return_pixel_mask" in kwargs:
             logger.warning_once(
@@ -1175,13 +1257,22 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         # prepare (COCO annotations as a list of Dict -> DETR target as a single Dict per image)
         if annotations is not None:
             prepared_images = []
             prepared_annotations = []
             for image, target in zip(images, annotations):
                 target = self.prepare_annotation(
-                    image, target, format, return_segmentation_masks=return_segmentation_masks, masks_path=masks_path
+                    image,
+                    target,
+                    format,
+                    return_segmentation_masks=return_segmentation_masks,
+                    masks_path=masks_path,
+                    input_data_format=input_data_format,
                 )
                 prepared_images.append(image)
                 prepared_annotations.append(target)
@@ -1194,33 +1285,49 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
             if annotations is not None:
                 resized_images, resized_annotations = [], []
                 for image, target in zip(images, annotations):
-                    orig_size = get_image_size(image)
-                    resized_image = self.resize(image, size=size, max_size=max_size, resample=resample)
-                    resized_annotation = self.resize_annotation(target, orig_size, get_image_size(resized_image))
+                    orig_size = get_image_size(image, input_data_format)
+                    resized_image = self.resize(
+                        image, size=size, max_size=max_size, resample=resample, input_data_format=input_data_format
+                    )
+                    resized_annotation = self.resize_annotation(
+                        target, orig_size, get_image_size(resized_image, input_data_format)
+                    )
                     resized_images.append(resized_image)
                     resized_annotations.append(resized_annotation)
                 images = resized_images
                 annotations = resized_annotations
                 del resized_images, resized_annotations
             else:
-                images = [self.resize(image, size=size, resample=resample) for image in images]
+                images = [
+                    self.resize(image, size=size, resample=resample, input_data_format=input_data_format)
+                    for image in images
+                ]
 
         if do_rescale:
-            images = [self.rescale(image, rescale_factor) for image in images]
+            images = [self.rescale(image, rescale_factor, input_data_format=input_data_format) for image in images]
 
         if do_normalize:
-            images = [self.normalize(image, image_mean, image_std) for image in images]
+            images = [
+                self.normalize(image, image_mean, image_std, input_data_format=input_data_format) for image in images
+            ]
             if annotations is not None:
                 annotations = [
-                    self.normalize_annotation(annotation, get_image_size(image))
+                    self.normalize_annotation(
+                        annotation, get_image_size(image, input_data_format), input_data_format=input_data_format
+                    )
                     for annotation, image in zip(annotations, images)
                 ]
 
         if do_pad:
             # Pads images and returns their mask: {'pixel_values': ..., 'pixel_mask': ...}
-            data = self.pad(images, return_pixel_mask=True, data_format=data_format)
+            data = self.pad(
+                images, return_pixel_mask=True, data_format=data_format, input_data_format=input_data_format
+            )
         else:
-            images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
+                for image in images
+            ]
             data = {"pixel_values": images}
 
         encoded_inputs = BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/deit/image_processing_deit.py

@@ -26,6 +26,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -114,6 +115,7 @@ class DeiTImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -131,6 +133,13 @@ class DeiTImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
 
         Returns:
             `np.ndarray`: The resized image.
@@ -139,7 +148,14 @@ class DeiTImageProcessor(BaseImageProcessor):
         if "height" not in size or "width" not in size:
             raise ValueError(f"The `size` dictionary must contain the keys `height` and `width`. Got {size.keys()}")
         output_size = (size["height"], size["width"])
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     def preprocess(
         self,
@@ -156,6 +172,7 @@ class DeiTImageProcessor(BaseImageProcessor):
         image_std: Optional[Union[float, List[float]]] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -197,6 +214,12 @@ class DeiTImageProcessor(BaseImageProcessor):
                 The channel dimension format for the output image. Can be one of:
                     - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                     - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         resample = resample if resample is not None else self.resample
@@ -235,19 +258,36 @@ class DeiTImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_center_crop:
-            images = [self.center_crop(image=image, size=crop_size) for image in images]
+            images = [
+                self.center_crop(image=image, size=crop_size, input_data_format=input_data_format) for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
-
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
+
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/deta/image_processing_deta.py

@@ -115,7 +115,10 @@ def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple[int, in
 
 # Copied from transformers.models.detr.image_processing_detr.get_resize_output_image_size
 def get_resize_output_image_size(
-    input_image: np.ndarray, size: Union[int, Tuple[int, int], List[int]], max_size: Optional[int] = None
+    input_image: np.ndarray,
+    size: Union[int, Tuple[int, int], List[int]],
+    max_size: Optional[int] = None,
+    input_data_format: Optional[Union[str, ChannelDimension]] = None,
 ) -> Tuple[int, int]:
     """
     Computes the output image size given the input image size and the desired output size. If the desired output size
@@ -129,8 +132,10 @@ def get_resize_output_image_size(
             The desired output size.
         max_size (`int`, *optional*):
             The maximum allowed output size.
+        input_data_format (`ChannelDimension` or `str`, *optional*):
+            The channel dimension format of the input image. If not provided, it will be inferred from the input image.
     """
-    image_size = get_image_size(input_image)
+    image_size = get_image_size(input_image, input_data_format)
     if isinstance(size, (list, tuple)):
         return size
 
@@ -200,23 +205,28 @@ def max_across_indices(values: Iterable[Any]) -> List[Any]:
 
 
 # Copied from transformers.models.detr.image_processing_detr.get_max_height_width
-def get_max_height_width(images: List[np.ndarray]) -> List[int]:
+def get_max_height_width(
+    images: List[np.ndarray], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> List[int]:
     """
     Get the maximum height and width across all images in a batch.
     """
-    input_channel_dimension = infer_channel_dimension_format(images[0])
+    if input_data_format is None:
+        input_data_format = infer_channel_dimension_format(images[0])
 
-    if input_channel_dimension == ChannelDimension.FIRST:
+    if input_data_format == ChannelDimension.FIRST:
         _, max_height, max_width = max_across_indices([img.shape for img in images])
-    elif input_channel_dimension == ChannelDimension.LAST:
+    elif input_data_format == ChannelDimension.LAST:
         max_height, max_width, _ = max_across_indices([img.shape for img in images])
     else:
-        raise ValueError(f"Invalid channel dimension format: {input_channel_dimension}")
+        raise ValueError(f"Invalid channel dimension format: {input_data_format}")
     return (max_height, max_width)
 
 
 # Copied from transformers.models.detr.image_processing_detr.make_pixel_mask
-def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarray:
+def make_pixel_mask(
+    image: np.ndarray, output_size: Tuple[int, int], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> np.ndarray:
     """
     Make a pixel mask for the image, where 1 indicates a valid pixel and 0 indicates padding.
 
@@ -226,7 +236,7 @@ def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarr
         output_size (`Tuple[int, int]`):
             Output size of the mask.
     """
-    input_height, input_width = get_image_size(image)
+    input_height, input_width = get_image_size(image, channel_dim=input_data_format)
     mask = np.zeros(output_size, dtype=np.int64)
     mask[:input_height, :input_width] = 1
     return mask
@@ -268,11 +278,16 @@ def convert_coco_poly_to_mask(segmentations, height: int, width: int) -> np.ndar
 
 
 # Copied from transformers.models.detr.image_processing_detr.prepare_coco_detection_annotation with DETR->DETA
-def prepare_coco_detection_annotation(image, target, return_segmentation_masks: bool = False):
+def prepare_coco_detection_annotation(
+    image,
+    target,
+    return_segmentation_masks: bool = False,
+    input_data_format: Optional[Union[ChannelDimension, str]] = None,
+):
     """
     Convert the target in COCO format into the format expected by DETA.
     """
-    image_height, image_width = get_image_size(image)
+    image_height, image_width = get_image_size(image, channel_dim=input_data_format)
 
     image_id = target["image_id"]
     image_id = np.asarray([image_id], dtype=np.int64)
@@ -357,12 +372,16 @@ def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
 
 # Copied from transformers.models.detr.image_processing_detr.prepare_coco_panoptic_annotation with DETR->DETA
 def prepare_coco_panoptic_annotation(
-    image: np.ndarray, target: Dict, masks_path: Union[str, pathlib.Path], return_masks: bool = True
+    image: np.ndarray,
+    target: Dict,
+    masks_path: Union[str, pathlib.Path],
+    return_masks: bool = True,
+    input_data_format: Union[ChannelDimension, str] = None,
 ) -> Dict:
     """
     Prepare a coco panoptic annotation for DETA.
     """
-    image_height, image_width = get_image_size(image)
+    image_height, image_width = get_image_size(image, channel_dim=input_data_format)
     annotation_path = pathlib.Path(masks_path) / target["file_name"]
 
     new_target = {}
@@ -522,6 +541,7 @@ class DetaImageProcessor(BaseImageProcessor):
         format: Optional[AnnotionFormat] = None,
         return_segmentation_masks: bool = None,
         masks_path: Optional[Union[str, pathlib.Path]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> Dict:
         """
         Prepare an annotation for feeding into DETA model.
@@ -530,11 +550,17 @@ class DetaImageProcessor(BaseImageProcessor):
 
         if format == AnnotionFormat.COCO_DETECTION:
             return_segmentation_masks = False if return_segmentation_masks is None else return_segmentation_masks
-            target = prepare_coco_detection_annotation(image, target, return_segmentation_masks)
+            target = prepare_coco_detection_annotation(
+                image, target, return_segmentation_masks, input_data_format=input_data_format
+            )
         elif format == AnnotionFormat.COCO_PANOPTIC:
             return_segmentation_masks = True if return_segmentation_masks is None else return_segmentation_masks
             target = prepare_coco_panoptic_annotation(
-                image, target, masks_path=masks_path, return_masks=return_segmentation_masks
+                image,
+                target,
+                masks_path=masks_path,
+                return_masks=return_segmentation_masks,
+                input_data_format=input_data_format,
             )
         else:
             raise ValueError(f"Format {format} is not supported.")
@@ -571,15 +597,32 @@ class DetaImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BILINEAR,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
         Resize the image to the given size. Size can be `min_size` (scalar) or `(height, width)` tuple. If size is an
         int, smaller edge of the image will be matched to this number.
+
+        Args:
+            image (`np.ndarray`):
+                Image to resize.
+            size (`Dict[str, int]`):
+                The desired output size. Can contain keys `shortest_edge` and `longest_edge` or `height` and `width`.
+            resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`):
+                Resampling filter to use if resizing the image.
+            data_format (`ChannelDimension`, *optional*):
+                The channel dimension format for the output image. If unset, the channel dimension format of the input
+                image is used.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred from the input
+                image.
         """
         size = get_size_dict(size, default_to_square=False)
         if "shortest_edge" in size and "longest_edge" in size:
-            size = get_resize_output_image_size(image, size["shortest_edge"], size["longest_edge"])
+            size = get_resize_output_image_size(
+                image, size["shortest_edge"], size["longest_edge"], input_data_format=input_data_format
+            )
         elif "height" in size and "width" in size:
             size = (size["height"], size["width"])
         else:
@@ -587,7 +630,9 @@ class DetaImageProcessor(BaseImageProcessor):
                 "Size must contain 'height' and 'width' keys or 'shortest_edge' and 'longest_edge' keys. Got"
                 f" {size.keys()}."
             )
-        image = resize(image, size=size, resample=resample, data_format=data_format)
+        image = resize(
+            image, size=size, resample=resample, data_format=data_format, input_data_format=input_data_format
+        )
         return image
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.resize_annotation
@@ -606,7 +651,11 @@ class DetaImageProcessor(BaseImageProcessor):
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.rescale
     def rescale(
-        self, image: np.ndarray, rescale_factor: float, data_format: Optional[Union[str, ChannelDimension]] = None
+        self,
+        image: np.ndarray,
+        rescale_factor: float,
+        data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Rescale the image by the given factor. image = image * rescale_factor.
@@ -621,8 +670,13 @@ class DetaImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the input image. If unset, is inferred from the input image. Can be
+                one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
         """
-        return rescale(image, rescale_factor, data_format=data_format)
+        return rescale(image, rescale_factor, data_format=data_format, input_data_format=input_data_format)
 
     # Copied from transformers.models.detr.image_processing_detr.DetrImageProcessor.normalize_annotation
     def normalize_annotation(self, annotation: Dict, image_size: Tuple[int, int]) -> Dict:
@@ -639,18 +693,24 @@ class DetaImageProcessor(BaseImageProcessor):
         output_size: Tuple[int, int],
         constant_values: Union[float, Iterable[float]] = 0,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Pad an image with zeros to the given size.
         """
-        input_height, input_width = get_image_size(image)
+        input_height, input_width = get_image_size(image, channel_dim=input_data_format)
         output_height, output_width = output_size
 
         pad_bottom = output_height - input_height
         pad_right = output_width - input_width
         padding = ((0, pad_bottom), (0, pad_right))
         padded_image = pad(
-            image, padding, mode=PaddingMode.CONSTANT, constant_values=constant_values, data_format=data_format
+            image,
+            padding,
+            mode=PaddingMode.CONSTANT,
+            constant_values=constant_values,
+            data_format=data_format,
+            input_data_format=input_data_format,
         )
         return padded_image
 
@@ -662,6 +722,7 @@ class DetaImageProcessor(BaseImageProcessor):
         return_pixel_mask: bool = True,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> BatchFeature:
         """
         Pads a batch of images to the bottom and right of the image with zeros to the size of largest height and width
@@ -683,17 +744,28 @@ class DetaImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
-        pad_size = get_max_height_width(images)
+        pad_size = get_max_height_width(images, input_data_format=input_data_format)
 
         padded_images = [
-            self._pad_image(image, pad_size, constant_values=constant_values, data_format=data_format)
+            self._pad_image(
+                image,
+                pad_size,
+                constant_values=constant_values,
+                data_format=data_format,
+                input_data_format=input_data_format,
+            )
             for image in images
         ]
         data = {"pixel_values": padded_images}
 
         if return_pixel_mask:
-            masks = [make_pixel_mask(image=image, output_size=pad_size) for image in images]
+            masks = [
+                make_pixel_mask(image=image, output_size=pad_size, input_data_format=input_data_format)
+                for image in images
+            ]
             data["pixel_mask"] = masks
 
         return BatchFeature(data=data, tensor_type=return_tensors)
@@ -716,6 +788,7 @@ class DetaImageProcessor(BaseImageProcessor):
         format: Optional[Union[str, AnnotionFormat]] = None,
         return_tensors: Optional[Union[TensorType, str]] = None,
         data_format: Union[str, ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> BatchFeature:
         """
@@ -761,8 +834,17 @@ class DetaImageProcessor(BaseImageProcessor):
                 Format of the annotations.
             return_tensors (`str` or `TensorType`, *optional*, defaults to self.return_tensors):
                 Type of tensors to return. If `None`, will return the list of images.
-            data_format (`str` or `ChannelDimension`, *optional*, defaults to self.data_format):
-                The channel dimension format of the image. If not provided, it will be the same as the input image.
+            data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
+                The channel dimension format for the output image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         if "pad_and_return_pixel_mask" in kwargs:
             logger.warning_once(
@@ -839,13 +921,22 @@ class DetaImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         # prepare (COCO annotations as a list of Dict -> DETR target as a single Dict per image)
         if annotations is not None:
             prepared_images = []
             prepared_annotations = []
             for image, target in zip(images, annotations):
                 target = self.prepare_annotation(
-                    image, target, format, return_segmentation_masks=return_segmentation_masks, masks_path=masks_path
+                    image,
+                    target,
+                    format,
+                    return_segmentation_masks=return_segmentation_masks,
+                    masks_path=masks_path,
+                    input_data_format=input_data_format,
                 )
                 prepared_images.append(image)
                 prepared_annotations.append(target)
@@ -858,33 +949,47 @@ class DetaImageProcessor(BaseImageProcessor):
             if annotations is not None:
                 resized_images, resized_annotations = [], []
                 for image, target in zip(images, annotations):
-                    orig_size = get_image_size(image)
-                    resized_image = self.resize(image, size=size, resample=resample)
-                    resized_annotation = self.resize_annotation(target, orig_size, get_image_size(resized_image))
+                    orig_size = get_image_size(image, input_data_format)
+                    resized_image = self.resize(
+                        image, size=size, resample=resample, input_data_format=input_data_format
+                    )
+                    resized_annotation = self.resize_annotation(
+                        target, orig_size, get_image_size(resized_image, input_data_format)
+                    )
                     resized_images.append(resized_image)
                     resized_annotations.append(resized_annotation)
                 images = resized_images
                 annotations = resized_annotations
                 del resized_images, resized_annotations
             else:
-                images = [self.resize(image, size=size, resample=resample) for image in images]
+                images = [
+                    self.resize(image, size=size, resample=resample, input_data_format=input_data_format)
+                    for image in images
+                ]
 
         if do_rescale:
-            images = [self.rescale(image, rescale_factor) for image in images]
+            images = [self.rescale(image, rescale_factor, input_data_format=input_data_format) for image in images]
 
         if do_normalize:
-            images = [self.normalize(image, image_mean, image_std) for image in images]
+            images = [
+                self.normalize(image, image_mean, image_std, input_data_format=input_data_format) for image in images
+            ]
             if annotations is not None:
                 annotations = [
-                    self.normalize_annotation(annotation, get_image_size(image))
+                    self.normalize_annotation(annotation, get_image_size(image, input_data_format))
                     for annotation, image in zip(annotations, images)
                 ]
 
         if do_pad:
             # Pads images and returns their mask: {'pixel_values': ..., 'pixel_mask': ...}
-            data = self.pad(images, return_pixel_mask=True, data_format=data_format)
+            data = self.pad(
+                images, return_pixel_mask=True, data_format=data_format, input_data_format=input_data_format
+            )
         else:
-            images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
+                for image in images
+            ]
             data = {"pixel_values": images}
 
         encoded_inputs = BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/detr/image_processing_detr.py

@@ -121,7 +121,10 @@ def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple[int, in
 
 
 def get_resize_output_image_size(
-    input_image: np.ndarray, size: Union[int, Tuple[int, int], List[int]], max_size: Optional[int] = None
+    input_image: np.ndarray,
+    size: Union[int, Tuple[int, int], List[int]],
+    max_size: Optional[int] = None,
+    input_data_format: Optional[Union[str, ChannelDimension]] = None,
 ) -> Tuple[int, int]:
     """
     Computes the output image size given the input image size and the desired output size. If the desired output size
@@ -135,8 +138,10 @@ def get_resize_output_image_size(
             The desired output size.
         max_size (`int`, *optional*):
             The maximum allowed output size.
+        input_data_format (`ChannelDimension` or `str`, *optional*):
+            The channel dimension format of the input image. If not provided, it will be inferred from the input image.
     """
-    image_size = get_image_size(input_image)
+    image_size = get_image_size(input_image, input_data_format)
     if isinstance(size, (list, tuple)):
         return size
 
@@ -203,23 +208,28 @@ def max_across_indices(values: Iterable[Any]) -> List[Any]:
 
 
 # Copied from transformers.models.vilt.image_processing_vilt.get_max_height_width
-def get_max_height_width(images: List[np.ndarray]) -> List[int]:
+def get_max_height_width(
+    images: List[np.ndarray], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> List[int]:
     """
     Get the maximum height and width across all images in a batch.
     """
-    input_channel_dimension = infer_channel_dimension_format(images[0])
+    if input_data_format is None:
+        input_data_format = infer_channel_dimension_format(images[0])
 
-    if input_channel_dimension == ChannelDimension.FIRST:
+    if input_data_format == ChannelDimension.FIRST:
         _, max_height, max_width = max_across_indices([img.shape for img in images])
-    elif input_channel_dimension == ChannelDimension.LAST:
+    elif input_data_format == ChannelDimension.LAST:
         max_height, max_width, _ = max_across_indices([img.shape for img in images])
     else:
-        raise ValueError(f"Invalid channel dimension format: {input_channel_dimension}")
+        raise ValueError(f"Invalid channel dimension format: {input_data_format}")
     return (max_height, max_width)
 
 
 # Copied from transformers.models.vilt.image_processing_vilt.make_pixel_mask
-def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarray:
+def make_pixel_mask(
+    image: np.ndarray, output_size: Tuple[int, int], input_data_format: Optional[Union[str, ChannelDimension]] = None
+) -> np.ndarray:
     """
     Make a pixel mask for the image, where 1 indicates a valid pixel and 0 indicates padding.
 
@@ -229,7 +239,7 @@ def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> np.ndarr
         output_size (`Tuple[int, int]`):
             Output size of the mask.
     """
-    input_height, input_width = get_image_size(image)
+    input_height, input_width = get_image_size(image, channel_dim=input_data_format)
     mask = np.zeros(output_size, dtype=np.int64)
     mask[:input_height, :input_width] = 1
     return mask
@@ -271,11 +281,16 @@ def convert_coco_poly_to_mask(segmentations, height: int, width: int) -> np.ndar
 
 
 # inspired by https://github.com/facebookresearch/detr/blob/master/datasets/coco.py#L50
-def prepare_coco_detection_annotation(image, target, return_segmentation_masks: bool = False):
+def prepare_coco_detection_annotation(
+    image,
+    target,
+    return_segmentation_masks: bool = False,
+    input_data_format: Optional[Union[ChannelDimension, str]] = None,
+):
     """
     Convert the target in COCO format into the format expected by DETR.
     """
-    image_height, image_width = get_image_size(image)
+    image_height, image_width = get_image_size(image, channel_dim=input_data_format)
 
     image_id = target["image_id"]
     image_id = np.asarray([image_id], dtype=np.int64)
@@ -358,12 +373,16 @@ def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
 
 
 def prepare_coco_panoptic_annotation(
-    image: np.ndarray, target: Dict, masks_path: Union[str, pathlib.Path], return_masks: bool = True
+    image: np.ndarray,
+    target: Dict,
+    masks_path: Union[str, pathlib.Path],
+    return_masks: bool = True,
+    input_data_format: Union[ChannelDimension, str] = None,
 ) -> Dict:
     """
     Prepare a coco panoptic annotation for DETR.
     """
-    image_height, image_width = get_image_size(image)
+    image_height, image_width = get_image_size(image, channel_dim=input_data_format)
     annotation_path = pathlib.Path(masks_path) / target["file_name"]
 
     new_target = {}
@@ -822,6 +841,7 @@ class DetrImageProcessor(BaseImageProcessor):
         format: Optional[AnnotionFormat] = None,
         return_segmentation_masks: bool = None,
         masks_path: Optional[Union[str, pathlib.Path]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> Dict:
         """
         Prepare an annotation for feeding into DETR model.
@@ -830,11 +850,17 @@ class DetrImageProcessor(BaseImageProcessor):
 
         if format == AnnotionFormat.COCO_DETECTION:
             return_segmentation_masks = False if return_segmentation_masks is None else return_segmentation_masks
-            target = prepare_coco_detection_annotation(image, target, return_segmentation_masks)
+            target = prepare_coco_detection_annotation(
+                image, target, return_segmentation_masks, input_data_format=input_data_format
+            )
         elif format == AnnotionFormat.COCO_PANOPTIC:
             return_segmentation_masks = True if return_segmentation_masks is None else return_segmentation_masks
             target = prepare_coco_panoptic_annotation(
-                image, target, masks_path=masks_path, return_masks=return_segmentation_masks
+                image,
+                target,
+                masks_path=masks_path,
+                return_masks=return_segmentation_masks,
+                input_data_format=input_data_format,
             )
         else:
             raise ValueError(f"Format {format} is not supported.")
@@ -867,11 +893,26 @@ class DetrImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BILINEAR,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
         Resize the image to the given size. Size can be `min_size` (scalar) or `(height, width)` tuple. If size is an
         int, smaller edge of the image will be matched to this number.
+
+        Args:
+            image (`np.ndarray`):
+                Image to resize.
+            size (`Dict[str, int]`):
+                Dictionary containing the size to resize to. Can contain the keys `shortest_edge` and `longest_edge` or
+                `height` and `width`.
+            resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`):
+                Resampling filter to use if resizing the image.
+            data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the output image. If unset, the channel dimension format of the input
+                image is used.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         if "max_size" in kwargs:
             logger.warning_once(
@@ -883,7 +924,9 @@ class DetrImageProcessor(BaseImageProcessor):
             max_size = None
         size = get_size_dict(size, max_size=max_size, default_to_square=False)
         if "shortest_edge" in size and "longest_edge" in size:
-            size = get_resize_output_image_size(image, size["shortest_edge"], size["longest_edge"])
+            size = get_resize_output_image_size(
+                image, size["shortest_edge"], size["longest_edge"], input_data_format=input_data_format
+            )
         elif "height" in size and "width" in size:
             size = (size["height"], size["width"])
         else:
@@ -891,7 +934,9 @@ class DetrImageProcessor(BaseImageProcessor):
                 "Size must contain 'height' and 'width' keys or 'shortest_edge' and 'longest_edge' keys. Got"
                 f" {size.keys()}."
             )
-        image = resize(image, size=size, resample=resample, data_format=data_format)
+        image = resize(
+            image, size=size, resample=resample, data_format=data_format, input_data_format=input_data_format, **kwargs
+        )
         return image
 
     def resize_annotation(
@@ -909,7 +954,11 @@ class DetrImageProcessor(BaseImageProcessor):
 
     # TODO (Amy) - update to use `rescale_factor` instead of `scale`
     def rescale(
-        self, image: np.ndarray, rescale_factor: float, data_format: Optional[Union[str, ChannelDimension]] = None
+        self,
+        image: np.ndarray,
+        rescale_factor: float,
+        data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Rescale the image by the given factor. image = image * rescale_factor.
@@ -924,8 +973,13 @@ class DetrImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the input image. If unset, is inferred from the input image. Can be
+                one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
         """
-        return rescale(image, rescale_factor, data_format=data_format)
+        return rescale(image, rescale_factor, data_format=data_format, input_data_format=input_data_format)
 
     def normalize_annotation(self, annotation: Dict, image_size: Tuple[int, int]) -> Dict:
         """
@@ -940,18 +994,24 @@ class DetrImageProcessor(BaseImageProcessor):
         output_size: Tuple[int, int],
         constant_values: Union[float, Iterable[float]] = 0,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Pad an image with zeros to the given size.
         """
-        input_height, input_width = get_image_size(image)
+        input_height, input_width = get_image_size(image, channel_dim=input_data_format)
         output_height, output_width = output_size
 
         pad_bottom = output_height - input_height
         pad_right = output_width - input_width
         padding = ((0, pad_bottom), (0, pad_right))
         padded_image = pad(
-            image, padding, mode=PaddingMode.CONSTANT, constant_values=constant_values, data_format=data_format
+            image,
+            padding,
+            mode=PaddingMode.CONSTANT,
+            constant_values=constant_values,
+            data_format=data_format,
+            input_data_format=input_data_format,
         )
         return padded_image
 
@@ -962,6 +1022,7 @@ class DetrImageProcessor(BaseImageProcessor):
         return_pixel_mask: bool = True,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> BatchFeature:
         """
         Pads a batch of images to the bottom and right of the image with zeros to the size of largest height and width
@@ -983,17 +1044,28 @@ class DetrImageProcessor(BaseImageProcessor):
                     - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
-        pad_size = get_max_height_width(images)
+        pad_size = get_max_height_width(images, input_data_format=input_data_format)
 
         padded_images = [
-            self._pad_image(image, pad_size, constant_values=constant_values, data_format=data_format)
+            self._pad_image(
+                image,
+                pad_size,
+                constant_values=constant_values,
+                data_format=data_format,
+                input_data_format=input_data_format,
+            )
             for image in images
         ]
         data = {"pixel_values": padded_images}
 
         if return_pixel_mask:
-            masks = [make_pixel_mask(image=image, output_size=pad_size) for image in images]
+            masks = [
+                make_pixel_mask(image=image, output_size=pad_size, input_data_format=input_data_format)
+                for image in images
+            ]
             data["pixel_mask"] = masks
 
         return BatchFeature(data=data, tensor_type=return_tensors)
@@ -1016,6 +1088,7 @@ class DetrImageProcessor(BaseImageProcessor):
         format: Optional[Union[str, AnnotionFormat]] = None,
         return_tensors: Optional[Union[TensorType, str]] = None,
         data_format: Union[str, ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> BatchFeature:
         """
@@ -1061,8 +1134,17 @@ class DetrImageProcessor(BaseImageProcessor):
                 Format of the annotations.
             return_tensors (`str` or `TensorType`, *optional*, defaults to self.return_tensors):
                 Type of tensors to return. If `None`, will return the list of images.
-            data_format (`str` or `ChannelDimension`, *optional*, defaults to self.data_format):
-                The channel dimension format of the image. If not provided, it will be the same as the input image.
+            data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
+                The channel dimension format for the output image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         if "pad_and_return_pixel_mask" in kwargs:
             logger.warning_once(
@@ -1147,13 +1229,22 @@ class DetrImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         # prepare (COCO annotations as a list of Dict -> DETR target as a single Dict per image)
         if annotations is not None:
             prepared_images = []
             prepared_annotations = []
             for image, target in zip(images, annotations):
                 target = self.prepare_annotation(
-                    image, target, format, return_segmentation_masks=return_segmentation_masks, masks_path=masks_path
+                    image,
+                    target,
+                    format,
+                    return_segmentation_masks=return_segmentation_masks,
+                    masks_path=masks_path,
+                    input_data_format=input_data_format,
                 )
                 prepared_images.append(image)
                 prepared_annotations.append(target)
@@ -1166,33 +1257,49 @@ class DetrImageProcessor(BaseImageProcessor):
             if annotations is not None:
                 resized_images, resized_annotations = [], []
                 for image, target in zip(images, annotations):
-                    orig_size = get_image_size(image)
-                    resized_image = self.resize(image, size=size, max_size=max_size, resample=resample)
-                    resized_annotation = self.resize_annotation(target, orig_size, get_image_size(resized_image))
+                    orig_size = get_image_size(image, input_data_format)
+                    resized_image = self.resize(
+                        image, size=size, max_size=max_size, resample=resample, input_data_format=input_data_format
+                    )
+                    resized_annotation = self.resize_annotation(
+                        target, orig_size, get_image_size(resized_image, input_data_format)
+                    )
                     resized_images.append(resized_image)
                     resized_annotations.append(resized_annotation)
                 images = resized_images
                 annotations = resized_annotations
                 del resized_images, resized_annotations
             else:
-                images = [self.resize(image, size=size, resample=resample) for image in images]
+                images = [
+                    self.resize(image, size=size, resample=resample, input_data_format=input_data_format)
+                    for image in images
+                ]
 
         if do_rescale:
-            images = [self.rescale(image, rescale_factor) for image in images]
+            images = [self.rescale(image, rescale_factor, input_data_format=input_data_format) for image in images]
 
         if do_normalize:
-            images = [self.normalize(image, image_mean, image_std) for image in images]
+            images = [
+                self.normalize(image, image_mean, image_std, input_data_format=input_data_format) for image in images
+            ]
             if annotations is not None:
                 annotations = [
-                    self.normalize_annotation(annotation, get_image_size(image))
+                    self.normalize_annotation(
+                        annotation, get_image_size(image, input_data_format), input_data_format=input_data_format
+                    )
                     for annotation, image in zip(annotations, images)
                 ]
 
         if do_pad:
             # Pads images and returns their mask: {'pixel_values': ..., 'pixel_mask': ...}
-            data = self.pad(images, return_pixel_mask=True, data_format=data_format)
+            data = self.pad(
+                images, return_pixel_mask=True, data_format=data_format, input_data_format=input_data_format
+            )
         else:
-            images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
+                for image in images
+            ]
             data = {"pixel_values": images}
 
         encoded_inputs = BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/donut/image_processing_donut.py

@@ -32,6 +32,7 @@ from ...image_utils import (
     ImageInput,
     PILImageResampling,
     get_image_size,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -122,7 +123,11 @@ class DonutImageProcessor(BaseImageProcessor):
         self.image_std = image_std if image_std is not None else IMAGENET_STANDARD_STD
 
     def align_long_axis(
-        self, image: np.ndarray, size: Dict[str, int], data_format: Optional[Union[str, ChannelDimension]] = None
+        self,
+        image: np.ndarray,
+        size: Dict[str, int],
+        data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Align the long axis of the image to the longest axis of the specified size.
@@ -132,11 +137,15 @@ class DonutImageProcessor(BaseImageProcessor):
                 The image to be aligned.
             size (`Dict[str, int]`):
                 The size `{"height": h, "width": w}` to align the long axis to.
+            data_format (`str` or `ChannelDimension`, *optional*):
+                The data format of the output image. If unset, the same format as the input image is used.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
 
         Returns:
             `np.ndarray`: The aligned image.
         """
-        input_height, input_width = get_image_size(image)
+        input_height, input_width = get_image_size(image, channel_dim=input_data_format)
         output_height, output_width = size["height"], size["width"]
 
         if (output_width < output_height and input_width > input_height) or (
@@ -145,7 +154,7 @@ class DonutImageProcessor(BaseImageProcessor):
             image = np.rot90(image, 3)
 
         if data_format is not None:
-            image = to_channel_dimension_format(image, data_format)
+            image = to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
 
         return image
 
@@ -155,6 +164,7 @@ class DonutImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         random_padding: bool = False,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Pad the image to the specified size.
@@ -168,9 +178,11 @@ class DonutImageProcessor(BaseImageProcessor):
                 Whether to use random padding or not.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The data format of the output image. If unset, the same format as the input image is used.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         output_height, output_width = size["height"], size["width"]
-        input_height, input_width = get_image_size(image)
+        input_height, input_width = get_image_size(image, channel_dim=input_data_format)
 
         delta_width = output_width - input_width
         delta_height = output_height - input_height
@@ -186,7 +198,7 @@ class DonutImageProcessor(BaseImageProcessor):
         pad_right = delta_width - pad_left
 
         padding = ((pad_top, pad_bottom), (pad_left, pad_right))
-        return pad(image, padding, data_format=data_format)
+        return pad(image, padding, data_format=data_format, input_data_format=input_data_format)
 
     def pad(self, *args, **kwargs):
         logger.info("pad is deprecated and will be removed in version 4.27. Please use pad_image instead.")
@@ -198,6 +210,7 @@ class DonutImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -213,8 +226,10 @@ class DonutImageProcessor(BaseImageProcessor):
                 The resampling filter to use.
             data_format (`Optional[Union[str, ChannelDimension]]`, *optional*):
                 The data format of the output image. If unset, the same format as the input image is used.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
-        input_height, input_width = get_image_size(image)
+        input_height, input_width = get_image_size(image, channel_dim=input_data_format)
         output_height, output_width = size["height"], size["width"]
 
         # We always resize to the smallest of either the input or output size.
@@ -230,7 +245,13 @@ class DonutImageProcessor(BaseImageProcessor):
             height = int(input_height * width / input_width)
 
         return resize(
-            image, size=(height, width), resample=resample, reducing_gap=2.0, data_format=data_format, **kwargs
+            image,
+            size=(height, width),
+            resample=resample,
+            reducing_gap=2.0,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
         )
 
     def resize(
@@ -239,6 +260,7 @@ class DonutImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -254,11 +276,22 @@ class DonutImageProcessor(BaseImageProcessor):
                 Resampling filter to use when resiizing the image.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         size = get_size_dict(size)
         shortest_edge = min(size["height"], size["width"])
-        output_size = get_resize_output_image_size(image, size=shortest_edge, default_to_square=False)
-        resized_image = resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        output_size = get_resize_output_image_size(
+            image, size=shortest_edge, default_to_square=False, input_data_format=input_data_format
+        )
+        resized_image = resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
         return resized_image
 
     def preprocess(
@@ -278,6 +311,7 @@ class DonutImageProcessor(BaseImageProcessor):
         image_std: Optional[Union[float, List[float]]] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -327,6 +361,12 @@ class DonutImageProcessor(BaseImageProcessor):
                 - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
                 - Unset: defaults to the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size = size if size is not None else self.size
@@ -367,25 +407,45 @@ class DonutImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_align_long_axis:
-            images = [self.align_long_axis(image, size=size) for image in images]
+            images = [self.align_long_axis(image, size=size, input_data_format=input_data_format) for image in images]
 
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_thumbnail:
-            images = [self.thumbnail(image=image, size=size) for image in images]
+            images = [self.thumbnail(image=image, size=size, input_data_format=input_data_format) for image in images]
 
         if do_pad:
-            images = [self.pad_image(image=image, size=size, random_padding=random_padding) for image in images]
+            images = [
+                self.pad_image(
+                    image=image, size=size, random_padding=random_padding, input_data_format=input_data_format
+                )
+                for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
-
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
+
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/dpt/image_processing_dpt.py

@@ -28,6 +28,7 @@ from ...image_utils import (
     ImageInput,
     PILImageResampling,
     get_image_size,
+    infer_channel_dimension_format,
     is_torch_available,
     is_torch_tensor,
     make_list_of_images,
@@ -48,7 +49,11 @@ logger = logging.get_logger(__name__)
 
 
 def get_resize_output_image_size(
-    input_image: np.ndarray, output_size: Union[int, Iterable[int]], keep_aspect_ratio: bool, multiple: int
+    input_image: np.ndarray,
+    output_size: Union[int, Iterable[int]],
+    keep_aspect_ratio: bool,
+    multiple: int,
+    input_data_format: Optional[Union[str, ChannelDimension]] = None,
 ) -> Tuple[int, int]:
     def constraint_to_multiple_of(val, multiple, min_val=0, max_val=None):
         x = round(val / multiple) * multiple
@@ -63,7 +68,7 @@ def get_resize_output_image_size(
 
     output_size = (output_size, output_size) if isinstance(output_size, int) else output_size
 
-    input_height, input_width = get_image_size(input_image)
+    input_height, input_width = get_image_size(input_image, input_data_format)
     output_height, output_width = output_size
 
     # determine new height and width
@@ -97,7 +102,7 @@ class DPTImageProcessor(BaseImageProcessor):
         keep_aspect_ratio (`bool`, *optional*, defaults to `False`):
             If `True`, the image is resized to the largest possible size such that the aspect ratio is preserved. Can
             be overidden by `keep_aspect_ratio` in `preprocess`.
-        ensure_multiple_of (`int`, *optional*, defaults to `1`):
+        ensure_multiple_of (`int`, *optional*, defaults to 1):
             If `do_resize` is `True`, the image is resized to a size that is a multiple of this value. Can be overidden
             by `ensure_multiple_of` in `preprocess`.
         resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`):
@@ -156,6 +161,7 @@ class DPTImageProcessor(BaseImageProcessor):
         ensure_multiple_of: int = 1,
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -170,7 +176,7 @@ class DPTImageProcessor(BaseImageProcessor):
                 Target size of the output image.
             keep_aspect_ratio (`bool`, *optional*, defaults to `False`):
                 If `True`, the image is resized to the largest possible size such that the aspect ratio is preserved.
-            ensure_multiple_of (`int`, *optional*, defaults to `1`):
+            ensure_multiple_of (`int`, *optional*, defaults to 1):
                 The image is resized to a size that is a multiple of this value.
             resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BICUBIC`):
                 Defines the resampling filter to use if resizing the image. Otherwise, the image is resized to size
@@ -179,6 +185,8 @@ class DPTImageProcessor(BaseImageProcessor):
                 Resampling filter to use when resiizing the image.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
         size = get_size_dict(size)
         if "height" not in size or "width" not in size:
@@ -188,8 +196,16 @@ class DPTImageProcessor(BaseImageProcessor):
             output_size=(size["height"], size["width"]),
             keep_aspect_ratio=keep_aspect_ratio,
             multiple=ensure_multiple_of,
+            input_data_format=input_data_format,
+        )
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
         )
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
 
     def preprocess(
         self,
@@ -206,6 +222,7 @@ class DPTImageProcessor(BaseImageProcessor):
         image_std: Optional[Union[float, List[float]]] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -249,6 +266,12 @@ class DPTImageProcessor(BaseImageProcessor):
                 The channel dimension format for the output image. Can be one of:
                     - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                     - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size = size if size is not None else self.size
@@ -282,16 +305,31 @@ class DPTImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
-
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
+
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/efficientformer/image_processing_efficientformer.py

@@ -30,6 +30,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     is_batched,
     to_numpy_array,
     valid_images,
@@ -116,6 +117,7 @@ class EfficientFormerImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BILINEAR,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -133,6 +135,8 @@ class EfficientFormerImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
 
         Returns:
             `np.ndarray`: The resized image.
@@ -140,13 +144,17 @@ class EfficientFormerImageProcessor(BaseImageProcessor):
         size = get_size_dict(size)
 
         if "shortest_edge" in size:
-            size = get_resize_output_image_size(image, size=size["shortest_edge"], default_to_square=False)
+            size = get_resize_output_image_size(
+                image, size=size["shortest_edge"], default_to_square=False, input_data_format=input_data_format
+            )
             # size = get_resize_output_image_size(image, size["shortest_edge"], size["longest_edge"])
         elif "height" in size and "width" in size:
             size = (size["height"], size["width"])
         else:
             raise ValueError(f"Size must contain 'height' and 'width' keys or 'shortest_edge' key. Got {size.keys()}")
-        return resize(image, size=size, resample=resample, data_format=data_format, **kwargs)
+        return resize(
+            image, size=size, resample=resample, data_format=data_format, input_data_format=input_data_format, **kwargs
+        )
 
     def preprocess(
         self,
@@ -163,6 +171,7 @@ class EfficientFormerImageProcessor(BaseImageProcessor):
         image_std: Optional[Union[float, List[float]]] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Union[str, ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> BatchFeature:
         """
@@ -205,6 +214,12 @@ class EfficientFormerImageProcessor(BaseImageProcessor):
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
                 - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         do_rescale = do_rescale if do_rescale is not None else self.do_rescale
@@ -241,19 +256,36 @@ class EfficientFormerImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size_dict, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size_dict, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_center_crop:
-            images = [self.center_crop(image=image, size=crop_size) for image in images]
+            images = [
+                self.center_crop(image=image, size=crop_size, input_data_format=input_data_format) for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor) for image in images]
+            images = [
+                self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
 
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/efficientnet/image_processing_efficientnet.py

@@ -26,6 +26,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -123,6 +124,7 @@ class EfficientNetImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.NEAREST,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -140,6 +142,13 @@ class EfficientNetImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
 
         Returns:
             `np.ndarray`: The resized image.
@@ -148,7 +157,14 @@ class EfficientNetImageProcessor(BaseImageProcessor):
         if "height" not in size or "width" not in size:
             raise ValueError(f"The `size` dictionary must contain the keys `height` and `width`. Got {size.keys()}")
         output_size = (size["height"], size["width"])
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     def rescale(
         self,
@@ -156,6 +172,7 @@ class EfficientNetImageProcessor(BaseImageProcessor):
         scale: Union[int, float],
         offset: bool = True,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ):
         """
@@ -177,8 +194,12 @@ class EfficientNetImageProcessor(BaseImageProcessor):
                 Whether to scale the image in both negative and positive directions.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
-        rescaled_image = rescale(image, scale=scale, data_format=data_format, **kwargs)
+        rescaled_image = rescale(
+            image, scale=scale, data_format=data_format, input_data_format=input_data_format, **kwargs
+        )
 
         if offset:
             rescaled_image = rescaled_image - 1
@@ -202,6 +223,7 @@ class EfficientNetImageProcessor(BaseImageProcessor):
         include_top: bool = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -247,6 +269,12 @@ class EfficientNetImageProcessor(BaseImageProcessor):
                 The channel dimension format for the output image. Can be one of:
                     - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                     - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         resample = resample if resample is not None else self.resample
@@ -287,22 +315,44 @@ class EfficientNetImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_center_crop:
-            images = [self.center_crop(image=image, size=crop_size) for image in images]
+            images = [
+                self.center_crop(image=image, size=crop_size, input_data_format=input_data_format) for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image=image, scale=rescale_factor, offset=rescale_offset) for image in images]
+            images = [
+                self.rescale(
+                    image=image, scale=rescale_factor, offset=rescale_offset, input_data_format=input_data_format
+                )
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]
+            images = [
+                self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if include_top:
-            images = [self.normalize(image=image, mean=[0, 0, 0], std=image_std) for image in images]
-
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                self.normalize(image=image, mean=0, std=image_std, input_data_format=input_data_format)
+                for image in images
+            ]
+
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/flava/image_processing_flava.py

@@ -29,6 +29,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -338,6 +339,7 @@ class FlavaImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BICUBIC,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -355,6 +357,13 @@ class FlavaImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
 
         Returns:
             `np.ndarray`: The resized image.
@@ -363,7 +372,14 @@ class FlavaImageProcessor(BaseImageProcessor):
         if "height" not in size or "width" not in size:
             raise ValueError(f"The `size` dictionary must contain the keys `height` and `width`. Got {size.keys()}")
         output_size = (size["height"], size["width"])
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     def map_pixels(self, image: np.ndarray) -> np.ndarray:
         return (1 - 2 * LOGIT_LAPLACE_EPS) * image + LOGIT_LAPLACE_EPS
@@ -383,6 +399,7 @@ class FlavaImageProcessor(BaseImageProcessor):
         image_std: Optional[Union[float, List[float]]] = None,
         do_map_pixels: bool = None,
         data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[ChannelDimension] = None,
     ) -> np.ndarray:
         """Preprocesses a single image."""
         if do_resize and size is None or resample is None:
@@ -397,23 +414,27 @@ class FlavaImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         image = to_numpy_array(image)
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(image)
+
         if do_resize:
-            image = self.resize(image=image, size=size, resample=resample)
+            image = self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
 
         if do_center_crop:
-            image = self.center_crop(image=image, size=crop_size)
+            image = self.center_crop(image=image, size=crop_size, input_data_format=input_data_format)
 
         if do_rescale:
-            image = self.rescale(image=image, scale=rescale_factor)
+            image = self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
 
         if do_normalize:
-            image = self.normalize(image=image, mean=image_mean, std=image_std)
+            image = self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
 
         if do_map_pixels:
             image = self.map_pixels(image)
 
         if data_format is not None:
-            image = to_channel_dimension_format(image, data_format)
+            image = to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
         return image
 
     def preprocess(
@@ -452,6 +473,7 @@ class FlavaImageProcessor(BaseImageProcessor):
         codebook_image_std: Optional[Iterable[float]] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -533,6 +555,12 @@ class FlavaImageProcessor(BaseImageProcessor):
                 The channel dimension format for the output image. Can be one of:
                     - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                     - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size = size if size is not None else self.size
@@ -615,6 +643,7 @@ class FlavaImageProcessor(BaseImageProcessor):
                 image_std=image_std,
                 do_map_pixels=False,
                 data_format=data_format,
+                input_data_format=input_data_format,
             )
             for img in images
         ]
@@ -636,6 +665,7 @@ class FlavaImageProcessor(BaseImageProcessor):
                     image_std=codebook_image_std,
                     do_map_pixels=codebook_do_map_pixels,
                     data_format=data_format,
+                    input_data_format=input_data_format,
                 )
                 for img in images
             ]

src/transformers/models/glpn/image_processing_glpn.py

@@ -25,6 +25,7 @@ from ...image_utils import (
     ChannelDimension,
     PILImageResampling,
     get_image_size,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -75,6 +76,7 @@ class GLPNImageProcessor(BaseImageProcessor):
         size_divisor: int,
         resample: PILImageResampling = PILImageResampling.BILINEAR,
         data_format: Optional[ChannelDimension] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -95,15 +97,27 @@ class GLPNImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not set, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
 
         Returns:
             `np.ndarray`: The resized image.
         """
-        height, width = get_image_size(image)
+        height, width = get_image_size(image, channel_dim=input_data_format)
         # Rounds the height and width down to the closest multiple of size_divisor
         new_h = height // size_divisor * size_divisor
         new_w = width // size_divisor * size_divisor
-        image = resize(image, (new_h, new_w), resample=resample, data_format=data_format, **kwargs)
+        image = resize(
+            image,
+            (new_h, new_w),
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
         return image
 
     def preprocess(
@@ -115,6 +129,7 @@ class GLPNImageProcessor(BaseImageProcessor):
         do_rescale: Optional[bool] = None,
         return_tensors: Optional[Union[TensorType, str]] = None,
         data_format: ChannelDimension = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> BatchFeature:
         """
@@ -144,6 +159,12 @@ class GLPNImageProcessor(BaseImageProcessor):
                 The channel dimension format for the output image. Can be one of:
                     - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                     - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         do_rescale = do_rescale if do_rescale is not None else self.do_rescale
@@ -161,13 +182,22 @@ class GLPNImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(img) for img in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image, size_divisor=size_divisor, resample=resample) for image in images]
+            images = [
+                self.resize(image, size_divisor=size_divisor, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_rescale:
-            images = [self.rescale(image, scale=1 / 255) for image in images]
+            images = [self.rescale(image, scale=1 / 255, input_data_format=input_data_format) for image in images]
 
-        images = [to_channel_dimension_format(image, data_format) for image in images]
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+        ]
 
         data = {"pixel_values": images}
         return BatchFeature(data=data, tensor_type=return_tensors)

src/transformers/models/imagegpt/image_processing_imagegpt.py

@@ -24,6 +24,7 @@ from ...image_utils import (
     ChannelDimension,
     ImageInput,
     PILImageResampling,
+    infer_channel_dimension_format,
     make_list_of_images,
     to_numpy_array,
     valid_images,
@@ -107,6 +108,7 @@ class ImageGPTImageProcessor(BaseImageProcessor):
         size: Dict[str, int],
         resample: PILImageResampling = PILImageResampling.BILINEAR,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> np.ndarray:
         """
@@ -124,6 +126,13 @@ class ImageGPTImageProcessor(BaseImageProcessor):
                 image is used. Can be one of:
                 - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                 - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
 
         Returns:
             `np.ndarray`: The resized image.
@@ -132,12 +141,20 @@ class ImageGPTImageProcessor(BaseImageProcessor):
         if "height" not in size or "width" not in size:
             raise ValueError(f"The `size` dictionary must contain the keys `height` and `width`. Got {size.keys()}")
         output_size = (size["height"], size["width"])
-        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)
+        return resize(
+            image,
+            size=output_size,
+            resample=resample,
+            data_format=data_format,
+            input_data_format=input_data_format,
+            **kwargs,
+        )
 
     def normalize(
         self,
         image: np.ndarray,
         data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
     ) -> np.ndarray:
         """
         Normalizes an images' pixel values to between [-1, 1].
@@ -147,8 +164,10 @@ class ImageGPTImageProcessor(BaseImageProcessor):
                 Image to normalize.
             data_format (`str` or `ChannelDimension`, *optional*):
                 The channel dimension format of the image. If not provided, it will be the same as the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format of the input image. If not provided, it will be inferred.
         """
-        image = rescale(image=image, scale=1 / 127.5, data_format=data_format)
+        image = rescale(image=image, scale=1 / 127.5, data_format=data_format, input_data_format=input_data_format)
         image = image - 1
         return image
 
@@ -163,6 +182,7 @@ class ImageGPTImageProcessor(BaseImageProcessor):
         clusters: Optional[Union[List[List[int]], np.ndarray]] = None,
         return_tensors: Optional[Union[str, TensorType]] = None,
         data_format: Optional[Union[str, ChannelDimension]] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
         **kwargs,
     ) -> PIL.Image.Image:
         """
@@ -197,6 +217,12 @@ class ImageGPTImageProcessor(BaseImageProcessor):
                     - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                     - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
                 Only has an effect if `do_color_quantize` is set to `False`.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
         """
         do_resize = do_resize if do_resize is not None else self.do_resize
         size = size if size is not None else self.size
@@ -224,14 +250,21 @@ class ImageGPTImageProcessor(BaseImageProcessor):
         # All transformations expect numpy arrays.
         images = [to_numpy_array(image) for image in images]
 
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = infer_channel_dimension_format(images[0])
+
         if do_resize:
-            images = [self.resize(image=image, size=size, resample=resample) for image in images]
+            images = [
+                self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format)
+                for image in images
+            ]
 
         if do_normalize:
-            images = [self.normalize(image=image) for image in images]
+            images = [self.normalize(image=image, input_data_format=input_data_format) for image in images]
 
         if do_color_quantize:
-            images = [to_channel_dimension_format(image, ChannelDimension.LAST) for image in images]
+            images = [to_channel_dimension_format(image, ChannelDimension.LAST, input_data_format) for image in images]
             # color quantize from (batch_size, height, width, 3) to (batch_size, height, width)
             images = np.array(images)
             images = color_quantize(images, clusters).reshape(images.shape[:-1])
@@ -243,7 +276,10 @@ class ImageGPTImageProcessor(BaseImageProcessor):
             # We need to convert back to a list of images to keep consistent behaviour across processors.
             images = list(images)
         else:
-            images = [to_channel_dimension_format(image, data_format) for image in images]
+            images = [
+                to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
+                for image in images
+            ]
 
         data = {"input_ids": images}
         return BatchFeature(data=data, tensor_type=return_tensors)