functional.py 60.6 KB
Newer Older
1
import math
2
3
import numbers
import warnings
4
from enum import Enum
5
from typing import List, Tuple, Any, Optional
6
7
8

import numpy as np
import torch
9
from PIL import Image
10
11
from torch import Tensor

12
13
14
15
16
try:
    import accimage
except ImportError:
    accimage = None

17
from ..utils import _log_api_usage_once
18
19
20
from . import functional_pil as F_pil
from . import functional_tensor as F_t

21

22
class InterpolationMode(Enum):
23
    """Interpolation modes
24
    Available interpolation methods are ``nearest``, ``bilinear``, ``bicubic``, ``box``, ``hamming``, and ``lanczos``.
25
    """
26

27
28
29
30
31
32
33
34
35
36
    NEAREST = "nearest"
    BILINEAR = "bilinear"
    BICUBIC = "bicubic"
    # For PIL compatibility
    BOX = "box"
    HAMMING = "hamming"
    LANCZOS = "lanczos"


# TODO: Once torchscript supports Enums with staticmethod
37
38
# this can be put into InterpolationMode as staticmethod
def _interpolation_modes_from_int(i: int) -> InterpolationMode:
39
    inverse_modes_mapping = {
40
41
42
43
44
45
        0: InterpolationMode.NEAREST,
        2: InterpolationMode.BILINEAR,
        3: InterpolationMode.BICUBIC,
        4: InterpolationMode.BOX,
        5: InterpolationMode.HAMMING,
        1: InterpolationMode.LANCZOS,
46
47
48
49
50
    }
    return inverse_modes_mapping[i]


pil_modes_mapping = {
51
52
53
54
55
56
    InterpolationMode.NEAREST: 0,
    InterpolationMode.BILINEAR: 2,
    InterpolationMode.BICUBIC: 3,
    InterpolationMode.BOX: 4,
    InterpolationMode.HAMMING: 5,
    InterpolationMode.LANCZOS: 1,
57
58
}

vfdev's avatar
vfdev committed
59
60
61
_is_pil_image = F_pil._is_pil_image


62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
def get_dimensions(img: Tensor) -> List[int]:
    """Returns the dimensions of an image as [channels, height, width].

    Args:
        img (PIL Image or Tensor): The image to be checked.

    Returns:
        List[int]: The image dimensions.
    """
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(get_dimensions)
    if isinstance(img, torch.Tensor):
        return F_t.get_dimensions(img)

    return F_pil.get_dimensions(img)


79
80
81
82
83
84
85
86
def get_image_size(img: Tensor) -> List[int]:
    """Returns the size of an image as [width, height].

    Args:
        img (PIL Image or Tensor): The image to be checked.

    Returns:
        List[int]: The image size.
vfdev's avatar
vfdev committed
87
    """
88
89
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(get_image_size)
vfdev's avatar
vfdev committed
90
    if isinstance(img, torch.Tensor):
91
        return F_t.get_image_size(img)
92

93
    return F_pil.get_image_size(img)
94

vfdev's avatar
vfdev committed
95

96
97
98
99
100
101
102
103
def get_image_num_channels(img: Tensor) -> int:
    """Returns the number of channels of an image.

    Args:
        img (PIL Image or Tensor): The image to be checked.

    Returns:
        int: The number of channels.
104
    """
105
106
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(get_image_num_channels)
107
    if isinstance(img, torch.Tensor):
108
        return F_t.get_image_num_channels(img)
109

110
    return F_pil.get_image_num_channels(img)
111
112


vfdev's avatar
vfdev committed
113
114
@torch.jit.unused
def _is_numpy(img: Any) -> bool:
115
116
117
    return isinstance(img, np.ndarray)


vfdev's avatar
vfdev committed
118
119
@torch.jit.unused
def _is_numpy_image(img: Any) -> bool:
120
    return img.ndim in {2, 3}
121
122
123
124


def to_tensor(pic):
    """Convert a ``PIL Image`` or ``numpy.ndarray`` to tensor.
125
    This function does not support torchscript.
126

127
    See :class:`~torchvision.transforms.ToTensor` for more details.
128
129
130
131
132
133
134

    Args:
        pic (PIL Image or numpy.ndarray): Image to be converted to tensor.

    Returns:
        Tensor: Converted image.
    """
135
136
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(to_tensor)
137
    if not (F_pil._is_pil_image(pic) or _is_numpy(pic)):
138
        raise TypeError(f"pic should be PIL Image or ndarray. Got {type(pic)}")
139

140
    if _is_numpy(pic) and not _is_numpy_image(pic):
141
        raise ValueError(f"pic should be 2/3 dimensional. Got {pic.ndim} dimensions.")
142

143
144
    default_float_dtype = torch.get_default_dtype()

145
146
    if isinstance(pic, np.ndarray):
        # handle numpy array
surgan12's avatar
surgan12 committed
147
148
149
        if pic.ndim == 2:
            pic = pic[:, :, None]

150
        img = torch.from_numpy(pic.transpose((2, 0, 1))).contiguous()
151
        # backward compatibility
152
        if isinstance(img, torch.ByteTensor):
153
            return img.to(dtype=default_float_dtype).div(255)
154
155
        else:
            return img
156
157

    if accimage is not None and isinstance(pic, accimage.Image):
158
        nppic = np.zeros([pic.channels, pic.height, pic.width], dtype=np.float32)
159
        pic.copyto(nppic)
160
        return torch.from_numpy(nppic).to(dtype=default_float_dtype)
161
162

    # handle PIL Image
163
164
    mode_to_nptype = {"I": np.int32, "I;16": np.int16, "F": np.float32}
    img = torch.from_numpy(np.array(pic, mode_to_nptype.get(pic.mode, np.uint8), copy=True))
165

166
    if pic.mode == "1":
167
        img = 255 * img
168
    img = img.view(pic.size[1], pic.size[0], len(pic.getbands()))
169
    # put it from HWC to CHW format
170
    img = img.permute((2, 0, 1)).contiguous()
171
    if isinstance(img, torch.ByteTensor):
172
        return img.to(dtype=default_float_dtype).div(255)
173
174
175
176
    else:
        return img


177
178
def pil_to_tensor(pic):
    """Convert a ``PIL Image`` to a tensor of the same type.
179
    This function does not support torchscript.
180

vfdev's avatar
vfdev committed
181
    See :class:`~torchvision.transforms.PILToTensor` for more details.
182

183
184
185
186
    .. note::

        A deep copy of the underlying array is performed.

187
188
189
190
191
192
    Args:
        pic (PIL Image): Image to be converted to tensor.

    Returns:
        Tensor: Converted image.
    """
193
194
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(pil_to_tensor)
195
    if not F_pil._is_pil_image(pic):
196
        raise TypeError(f"pic should be PIL Image. Got {type(pic)}")
197
198

    if accimage is not None and isinstance(pic, accimage.Image):
199
200
        # accimage format is always uint8 internally, so always return uint8 here
        nppic = np.zeros([pic.channels, pic.height, pic.width], dtype=np.uint8)
201
202
203
204
        pic.copyto(nppic)
        return torch.as_tensor(nppic)

    # handle PIL Image
205
    img = torch.as_tensor(np.array(pic, copy=True))
206
207
208
209
210
211
    img = img.view(pic.size[1], pic.size[0], len(pic.getbands()))
    # put it from HWC to CHW format
    img = img.permute((2, 0, 1))
    return img


212
213
def convert_image_dtype(image: torch.Tensor, dtype: torch.dtype = torch.float) -> torch.Tensor:
    """Convert a tensor image to the given ``dtype`` and scale the values accordingly
214
    This function does not support PIL Image.
215
216
217
218
219
220

    Args:
        image (torch.Tensor): Image to be converted
        dtype (torch.dtype): Desired data type of the output

    Returns:
vfdev's avatar
vfdev committed
221
        Tensor: Converted image
222
223
224
225
226
227
228
229
230
231
232
233

    .. note::

        When converting from a smaller to a larger integer ``dtype`` the maximum values are **not** mapped exactly.
        If converted back and forth, this mismatch has no effect.

    Raises:
        RuntimeError: When trying to cast :class:`torch.float32` to :class:`torch.int32` or :class:`torch.int64` as
            well as for trying to cast :class:`torch.float64` to :class:`torch.int64`. These conversions might lead to
            overflow errors since the floating point ``dtype`` cannot store consecutive integers over the whole range
            of the integer ``dtype``.
    """
234
235
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(convert_image_dtype)
236
    if not isinstance(image, torch.Tensor):
237
        raise TypeError("Input img should be Tensor Image")
238
239

    return F_t.convert_image_dtype(image, dtype)
240
241


242
def to_pil_image(pic, mode=None):
243
    """Convert a tensor or an ndarray to PIL Image. This function does not support torchscript.
244

245
    See :class:`~torchvision.transforms.ToPILImage` for more details.
246
247
248
249
250

    Args:
        pic (Tensor or numpy.ndarray): Image to be converted to PIL Image.
        mode (`PIL.Image mode`_): color space and pixel depth of input data (optional).

251
    .. _PIL.Image mode: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes
252
253
254
255

    Returns:
        PIL Image: Image converted to PIL Image.
    """
256
257
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(to_pil_image)
258
    if not (isinstance(pic, torch.Tensor) or isinstance(pic, np.ndarray)):
259
        raise TypeError(f"pic should be Tensor or ndarray. Got {type(pic)}.")
260

Varun Agrawal's avatar
Varun Agrawal committed
261
262
    elif isinstance(pic, torch.Tensor):
        if pic.ndimension() not in {2, 3}:
263
            raise ValueError(f"pic should be 2/3 dimensional. Got {pic.ndimension()} dimensions.")
Varun Agrawal's avatar
Varun Agrawal committed
264
265
266

        elif pic.ndimension() == 2:
            # if 2D image, add channel dimension (CHW)
Surgan Jandial's avatar
Surgan Jandial committed
267
            pic = pic.unsqueeze(0)
Varun Agrawal's avatar
Varun Agrawal committed
268

269
270
        # check number of channels
        if pic.shape[-3] > 4:
271
            raise ValueError(f"pic should not have > 4 channels. Got {pic.shape[-3]} channels.")
272

Varun Agrawal's avatar
Varun Agrawal committed
273
274
    elif isinstance(pic, np.ndarray):
        if pic.ndim not in {2, 3}:
275
            raise ValueError(f"pic should be 2/3 dimensional. Got {pic.ndim} dimensions.")
Varun Agrawal's avatar
Varun Agrawal committed
276
277
278
279
280

        elif pic.ndim == 2:
            # if 2D image, add channel dimension (HWC)
            pic = np.expand_dims(pic, 2)

281
282
        # check number of channels
        if pic.shape[-1] > 4:
283
            raise ValueError(f"pic should not have > 4 channels. Got {pic.shape[-1]} channels.")
284

285
    npimg = pic
Varun Agrawal's avatar
Varun Agrawal committed
286
    if isinstance(pic, torch.Tensor):
287
        if pic.is_floating_point() and mode != "F":
288
289
            pic = pic.mul(255).byte()
        npimg = np.transpose(pic.cpu().numpy(), (1, 2, 0))
290
291

    if not isinstance(npimg, np.ndarray):
292
        raise TypeError("Input pic must be a torch.Tensor or NumPy ndarray, not {type(npimg)}")
293
294
295
296
297

    if npimg.shape[2] == 1:
        expected_mode = None
        npimg = npimg[:, :, 0]
        if npimg.dtype == np.uint8:
298
            expected_mode = "L"
vfdev's avatar
vfdev committed
299
        elif npimg.dtype == np.int16:
300
            expected_mode = "I;16"
vfdev's avatar
vfdev committed
301
        elif npimg.dtype == np.int32:
302
            expected_mode = "I"
303
        elif npimg.dtype == np.float32:
304
            expected_mode = "F"
305
        if mode is not None and mode != expected_mode:
306
            raise ValueError(f"Incorrect mode ({mode}) supplied for input type {np.dtype}. Should be {expected_mode}")
307
308
        mode = expected_mode

surgan12's avatar
surgan12 committed
309
    elif npimg.shape[2] == 2:
310
        permitted_2_channel_modes = ["LA"]
surgan12's avatar
surgan12 committed
311
        if mode is not None and mode not in permitted_2_channel_modes:
312
            raise ValueError(f"Only modes {permitted_2_channel_modes} are supported for 2D inputs")
surgan12's avatar
surgan12 committed
313
314

        if mode is None and npimg.dtype == np.uint8:
315
            mode = "LA"
surgan12's avatar
surgan12 committed
316

317
    elif npimg.shape[2] == 4:
318
        permitted_4_channel_modes = ["RGBA", "CMYK", "RGBX"]
319
        if mode is not None and mode not in permitted_4_channel_modes:
320
            raise ValueError(f"Only modes {permitted_4_channel_modes} are supported for 4D inputs")
321
322

        if mode is None and npimg.dtype == np.uint8:
323
            mode = "RGBA"
324
    else:
325
        permitted_3_channel_modes = ["RGB", "YCbCr", "HSV"]
326
        if mode is not None and mode not in permitted_3_channel_modes:
327
            raise ValueError(f"Only modes {permitted_3_channel_modes} are supported for 3D inputs")
328
        if mode is None and npimg.dtype == np.uint8:
329
            mode = "RGB"
330
331

    if mode is None:
332
        raise TypeError(f"Input type {npimg.dtype} is not supported")
333
334
335
336

    return Image.fromarray(npimg, mode=mode)


337
def normalize(tensor: Tensor, mean: List[float], std: List[float], inplace: bool = False) -> Tensor:
338
    """Normalize a float tensor image with mean and standard deviation.
339
    This transform does not support PIL Image.
340

341
    .. note::
surgan12's avatar
surgan12 committed
342
        This transform acts out of place by default, i.e., it does not mutates the input tensor.
343

344
    See :class:`~torchvision.transforms.Normalize` for more details.
345
346

    Args:
347
        tensor (Tensor): Float tensor image of size (C, H, W) or (B, C, H, W) to be normalized.
348
        mean (sequence): Sequence of means for each channel.
349
        std (sequence): Sequence of standard deviations for each channel.
350
        inplace(bool,optional): Bool to make this operation inplace.
351
352
353
354

    Returns:
        Tensor: Normalized Tensor image.
    """
355
356
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(normalize)
357
    if not isinstance(tensor, torch.Tensor):
358
        raise TypeError(f"img should be Tensor Image. Got {type(tensor)}")
359

360
    return F_t.normalize(tensor, mean=mean, std=std, inplace=inplace)
361
362


363
364
365
366
367
368
369
def resize(
    img: Tensor,
    size: List[int],
    interpolation: InterpolationMode = InterpolationMode.BILINEAR,
    max_size: Optional[int] = None,
    antialias: Optional[bool] = None,
) -> Tensor:
vfdev's avatar
vfdev committed
370
    r"""Resize the input image to the given size.
371
    If the image is torch Tensor, it is expected
vfdev's avatar
vfdev committed
372
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions
373

374
375
376
377
    .. warning::
        The output image might be different depending on its type: when downsampling, the interpolation of PIL images
        and tensors is slightly different, because PIL applies antialiasing. This may lead to significant differences
        in the performance of a network. Therefore, it is preferable to train and serve a model with the same input
378
379
        types. See also below the ``antialias`` parameter, which can help making the output of PIL images and tensors
        closer.
380

381
    Args:
vfdev's avatar
vfdev committed
382
        img (PIL Image or Tensor): Image to be resized.
383
384
        size (sequence or int): Desired output size. If size is a sequence like
            (h, w), the output size will be matched to this. If size is an int,
Vitaliy Chiley's avatar
Vitaliy Chiley committed
385
            the smaller edge of the image will be matched to this number maintaining
386
            the aspect ratio. i.e, if height > width, then image will be rescaled to
vfdev's avatar
vfdev committed
387
            :math:`\left(\text{size} \times \frac{\text{height}}{\text{width}}, \text{size}\right)`.
388
389
390

            .. note::
                In torchscript mode size as single int is not supported, use a sequence of length 1: ``[size, ]``.
391
392
393
394
        interpolation (InterpolationMode): Desired interpolation enum defined by
            :class:`torchvision.transforms.InterpolationMode`.
            Default is ``InterpolationMode.BILINEAR``. If input is Tensor, only ``InterpolationMode.NEAREST``,
            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.
395
            For backward compatibility integer values (e.g. ``PIL.Image.NEAREST``) are still acceptable.
396
397
398
399
        max_size (int, optional): The maximum allowed for the longer edge of
            the resized image: if the longer edge of the image is greater
            than ``max_size`` after being resized according to ``size``, then
            the image is resized again so that the longer edge is equal to
400
            ``max_size``. As a result, ``size`` might be overruled, i.e the
401
402
403
            smaller edge may be shorter than ``size``. This is only supported
            if ``size`` is an int (or a sequence of length 1 in torchscript
            mode).
404
        antialias (bool, optional): antialias flag. If ``img`` is PIL Image, the flag is ignored and anti-alias
405
406
407
            is always used. If ``img`` is Tensor, the flag is False by default and can be set to True for
            ``InterpolationMode.BILINEAR`` only mode. This can help making the output for PIL images and tensors
            closer.
408
409
410

            .. warning::
                There is no autodiff support for ``antialias=True`` option with input ``img`` as Tensor.
411
412

    Returns:
vfdev's avatar
vfdev committed
413
        PIL Image or Tensor: Resized image.
414
    """
415
416
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(resize)
417
418
419
    # Backward compatibility with integer value
    if isinstance(interpolation, int):
        warnings.warn(
420
421
            "Argument interpolation should be of type InterpolationMode instead of int. "
            "Please, use InterpolationMode enum."
422
423
424
        )
        interpolation = _interpolation_modes_from_int(interpolation)

425
426
    if not isinstance(interpolation, InterpolationMode):
        raise TypeError("Argument interpolation should be a InterpolationMode")
427

vfdev's avatar
vfdev committed
428
    if not isinstance(img, torch.Tensor):
429
        if antialias is not None and not antialias:
430
            warnings.warn("Anti-alias option is always applied for PIL Image input. Argument antialias is ignored.")
431
        pil_interpolation = pil_modes_mapping[interpolation]
432
        return F_pil.resize(img, size=size, interpolation=pil_interpolation, max_size=max_size)
vfdev's avatar
vfdev committed
433

434
    return F_t.resize(img, size=size, interpolation=interpolation.value, max_size=max_size, antialias=antialias)
435
436


437
438
def pad(img: Tensor, padding: List[int], fill: int = 0, padding_mode: str = "constant") -> Tensor:
    r"""Pad the given image on all sides with the given "pad" value.
439
    If the image is torch Tensor, it is expected
440
441
442
    to have [..., H, W] shape, where ... means at most 2 leading dimensions for mode reflect and symmetric,
    at most 3 leading dimensions for mode edge,
    and an arbitrary number of leading dimensions for mode constant
443
444

    Args:
445
        img (PIL Image or Tensor): Image to be padded.
446
447
448
        padding (int or sequence): Padding on each border. If a single int is provided this
            is used to pad all borders. If sequence of length 2 is provided this is the padding
            on left/right and top/bottom respectively. If a sequence of length 4 is provided
449
            this is the padding for the left, top, right and bottom borders respectively.
450
451
452
453

            .. note::
                In torchscript mode padding as single int is not supported, use a sequence of
                length 1: ``[padding, ]``.
454
455
456
457
458
        fill (number or str or tuple): Pixel fill value for constant fill. Default is 0.
            If a tuple of length 3, it is used to fill R, G, B channels respectively.
            This value is only used when the padding_mode is constant.
            Only number is supported for torch Tensor.
            Only int or str or tuple value is supported for PIL Image.
459
460
        padding_mode (str): Type of padding. Should be: constant, edge, reflect or symmetric.
            Default is constant.
461
462
463

            - constant: pads with a constant value, this value is specified with fill

464
465
            - edge: pads with the last value at the edge of the image.
              If input a 5D torch Tensor, the last 3 dimensions will be padded instead of the last 2
466

467
468
469
            - reflect: pads with reflection of image without repeating the last value on the edge.
              For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode
              will result in [3, 2, 1, 2, 3, 4, 3, 2]
470

471
472
473
            - symmetric: pads with reflection of image repeating the last value on the edge.
              For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode
              will result in [2, 1, 1, 2, 3, 4, 4, 3]
474
475

    Returns:
476
        PIL Image or Tensor: Padded image.
477
    """
478
479
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(pad)
480
481
    if not isinstance(img, torch.Tensor):
        return F_pil.pad(img, padding=padding, fill=fill, padding_mode=padding_mode)
482

483
    return F_t.pad(img, padding=padding, fill=fill, padding_mode=padding_mode)
484
485


vfdev's avatar
vfdev committed
486
487
def crop(img: Tensor, top: int, left: int, height: int, width: int) -> Tensor:
    """Crop the given image at specified location and output size.
488
    If the image is torch Tensor, it is expected
489
490
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
    If image size is smaller than output size along any edge, image is padded with 0 and then cropped.
491

492
    Args:
vfdev's avatar
vfdev committed
493
        img (PIL Image or Tensor): Image to be cropped. (0,0) denotes the top left corner of the image.
494
495
496
497
        top (int): Vertical component of the top left corner of the crop box.
        left (int): Horizontal component of the top left corner of the crop box.
        height (int): Height of the crop box.
        width (int): Width of the crop box.
498

499
    Returns:
vfdev's avatar
vfdev committed
500
        PIL Image or Tensor: Cropped image.
501
502
    """

503
504
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(crop)
vfdev's avatar
vfdev committed
505
506
    if not isinstance(img, torch.Tensor):
        return F_pil.crop(img, top, left, height, width)
507

vfdev's avatar
vfdev committed
508
    return F_t.crop(img, top, left, height, width)
509

vfdev's avatar
vfdev committed
510
511
512

def center_crop(img: Tensor, output_size: List[int]) -> Tensor:
    """Crops the given image at the center.
513
    If the image is torch Tensor, it is expected
514
515
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
    If image size is smaller than output size along any edge, image is padded with 0 and then center cropped.
516

517
    Args:
vfdev's avatar
vfdev committed
518
        img (PIL Image or Tensor): Image to be cropped.
519
        output_size (sequence or int): (height, width) of the crop box. If int or sequence with single int,
vfdev's avatar
vfdev committed
520
521
            it is used for both directions.

522
    Returns:
vfdev's avatar
vfdev committed
523
        PIL Image or Tensor: Cropped image.
524
    """
525
526
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(center_crop)
527
528
    if isinstance(output_size, numbers.Number):
        output_size = (int(output_size), int(output_size))
vfdev's avatar
vfdev committed
529
530
531
    elif isinstance(output_size, (tuple, list)) and len(output_size) == 1:
        output_size = (output_size[0], output_size[0])

532
    _, image_height, image_width = get_dimensions(img)
533
    crop_height, crop_width = output_size
vfdev's avatar
vfdev committed
534

535
536
537
538
539
540
541
542
    if crop_width > image_width or crop_height > image_height:
        padding_ltrb = [
            (crop_width - image_width) // 2 if crop_width > image_width else 0,
            (crop_height - image_height) // 2 if crop_height > image_height else 0,
            (crop_width - image_width + 1) // 2 if crop_width > image_width else 0,
            (crop_height - image_height + 1) // 2 if crop_height > image_height else 0,
        ]
        img = pad(img, padding_ltrb, fill=0)  # PIL uses fill value 0
543
        _, image_height, image_width = get_dimensions(img)
544
545
546
        if crop_width == image_width and crop_height == image_height:
            return img

547
548
    crop_top = int(round((image_height - crop_height) / 2.0))
    crop_left = int(round((image_width - crop_width) / 2.0))
549
    return crop(img, crop_top, crop_left, crop_height, crop_width)
550
551


552
def resized_crop(
553
554
555
556
557
558
559
    img: Tensor,
    top: int,
    left: int,
    height: int,
    width: int,
    size: List[int],
    interpolation: InterpolationMode = InterpolationMode.BILINEAR,
560
561
) -> Tensor:
    """Crop the given image and resize it to desired size.
562
    If the image is torch Tensor, it is expected
563
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions
564

565
    Notably used in :class:`~torchvision.transforms.RandomResizedCrop`.
566
567

    Args:
568
        img (PIL Image or Tensor): Image to be cropped. (0,0) denotes the top left corner of the image.
569
570
571
572
        top (int): Vertical component of the top left corner of the crop box.
        left (int): Horizontal component of the top left corner of the crop box.
        height (int): Height of the crop box.
        width (int): Width of the crop box.
573
        size (sequence or int): Desired output size. Same semantics as ``resize``.
574
575
576
577
        interpolation (InterpolationMode): Desired interpolation enum defined by
            :class:`torchvision.transforms.InterpolationMode`.
            Default is ``InterpolationMode.BILINEAR``. If input is Tensor, only ``InterpolationMode.NEAREST``,
            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.
578
579
            For backward compatibility integer values (e.g. ``PIL.Image.NEAREST``) are still acceptable.

580
    Returns:
581
        PIL Image or Tensor: Cropped image.
582
    """
583
584
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(resized_crop)
585
    img = crop(img, top, left, height, width)
586
587
588
589
    img = resize(img, size, interpolation)
    return img


590
def hflip(img: Tensor) -> Tensor:
591
    """Horizontally flip the given image.
592
593

    Args:
vfdev's avatar
vfdev committed
594
        img (PIL Image or Tensor): Image to be flipped. If img
595
            is a Tensor, it is expected to be in [..., H, W] format,
596
            where ... means it can have an arbitrary number of leading
597
            dimensions.
598
599

    Returns:
vfdev's avatar
vfdev committed
600
        PIL Image or Tensor:  Horizontally flipped image.
601
    """
602
603
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(hflip)
604
605
    if not isinstance(img, torch.Tensor):
        return F_pil.hflip(img)
606

607
    return F_t.hflip(img)
608
609


610
def _get_perspective_coeffs(startpoints: List[List[int]], endpoints: List[List[int]]) -> List[float]:
611
612
    """Helper function to get the coefficients (a, b, c, d, e, f, g, h) for the perspective transforms.

Vitaliy Chiley's avatar
Vitaliy Chiley committed
613
    In Perspective Transform each pixel (x, y) in the original image gets transformed as,
614
615
616
     (x, y) -> ( (ax + by + c) / (gx + hy + 1), (dx + ey + f) / (gx + hy + 1) )

    Args:
617
618
619
620
621
        startpoints (list of list of ints): List containing four lists of two integers corresponding to four corners
            ``[top-left, top-right, bottom-right, bottom-left]`` of the original image.
        endpoints (list of list of ints): List containing four lists of two integers corresponding to four corners
            ``[top-left, top-right, bottom-right, bottom-left]`` of the transformed image.

622
623
624
    Returns:
        octuple (a, b, c, d, e, f, g, h) for transforming each pixel.
    """
625
626
627
628
629
    a_matrix = torch.zeros(2 * len(startpoints), 8, dtype=torch.float)

    for i, (p1, p2) in enumerate(zip(endpoints, startpoints)):
        a_matrix[2 * i, :] = torch.tensor([p1[0], p1[1], 1, 0, 0, 0, -p2[0] * p1[0], -p2[0] * p1[1]])
        a_matrix[2 * i + 1, :] = torch.tensor([0, 0, 0, p1[0], p1[1], 1, -p2[1] * p1[0], -p2[1] * p1[1]])
630

631
    b_matrix = torch.tensor(startpoints, dtype=torch.float).view(8)
632
    res = torch.linalg.lstsq(a_matrix, b_matrix, driver="gels").solution
633

634
    output: List[float] = res.tolist()
635
    return output
636
637


638
def perspective(
639
640
641
642
643
    img: Tensor,
    startpoints: List[List[int]],
    endpoints: List[List[int]],
    interpolation: InterpolationMode = InterpolationMode.BILINEAR,
    fill: Optional[List[float]] = None,
644
645
) -> Tensor:
    """Perform perspective transform of the given image.
646
    If the image is torch Tensor, it is expected
647
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
648
649

    Args:
650
651
652
653
654
        img (PIL Image or Tensor): Image to be transformed.
        startpoints (list of list of ints): List containing four lists of two integers corresponding to four corners
            ``[top-left, top-right, bottom-right, bottom-left]`` of the original image.
        endpoints (list of list of ints): List containing four lists of two integers corresponding to four corners
            ``[top-left, top-right, bottom-right, bottom-left]`` of the transformed image.
655
656
657
        interpolation (InterpolationMode): Desired interpolation enum defined by
            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.
            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
658
            For backward compatibility integer values (e.g. ``PIL.Image.NEAREST``) are still acceptable.
659
660
        fill (sequence or number, optional): Pixel fill value for the area outside the transformed
            image. If given a number, the value is used for all bands respectively.
661
662
663
664

            .. note::
                In torchscript mode single int/float value is not supported, please use a sequence
                of length 1: ``[value, ]``.
665

666
    Returns:
667
        PIL Image or Tensor: transformed Image.
668
    """
669
670
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(perspective)
671

672
    coeffs = _get_perspective_coeffs(startpoints, endpoints)
673

674
675
676
    # Backward compatibility with integer value
    if isinstance(interpolation, int):
        warnings.warn(
677
678
            "Argument interpolation should be of type InterpolationMode instead of int. "
            "Please, use InterpolationMode enum."
679
680
681
        )
        interpolation = _interpolation_modes_from_int(interpolation)

682
683
    if not isinstance(interpolation, InterpolationMode):
        raise TypeError("Argument interpolation should be a InterpolationMode")
684

685
    if not isinstance(img, torch.Tensor):
686
687
        pil_interpolation = pil_modes_mapping[interpolation]
        return F_pil.perspective(img, coeffs, interpolation=pil_interpolation, fill=fill)
688

689
    return F_t.perspective(img, coeffs, interpolation=interpolation.value, fill=fill)
690
691


692
def vflip(img: Tensor) -> Tensor:
693
    """Vertically flip the given image.
694
695

    Args:
vfdev's avatar
vfdev committed
696
        img (PIL Image or Tensor): Image to be flipped. If img
697
            is a Tensor, it is expected to be in [..., H, W] format,
698
            where ... means it can have an arbitrary number of leading
699
            dimensions.
700
701

    Returns:
702
        PIL Image or Tensor:  Vertically flipped image.
703
    """
704
705
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(vflip)
706
707
    if not isinstance(img, torch.Tensor):
        return F_pil.vflip(img)
708

709
    return F_t.vflip(img)
710
711


vfdev's avatar
vfdev committed
712
713
def five_crop(img: Tensor, size: List[int]) -> Tuple[Tensor, Tensor, Tensor, Tensor, Tensor]:
    """Crop the given image into four corners and the central crop.
714
    If the image is torch Tensor, it is expected
vfdev's avatar
vfdev committed
715
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions
716
717
718
719
720
721

    .. Note::
        This transform returns a tuple of images and there may be a
        mismatch in the number of inputs and targets your ``Dataset`` returns.

    Args:
vfdev's avatar
vfdev committed
722
723
724
        img (PIL Image or Tensor): Image to be cropped.
        size (sequence or int): Desired output size of the crop. If size is an
            int instead of sequence like (h, w), a square crop (size, size) is
725
            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).
726

727
    Returns:
728
       tuple: tuple (tl, tr, bl, br, center)
729
       Corresponding top left, top right, bottom left, bottom right and center crop.
730
    """
731
732
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(five_crop)
733
734
    if isinstance(size, numbers.Number):
        size = (int(size), int(size))
vfdev's avatar
vfdev committed
735
736
    elif isinstance(size, (tuple, list)) and len(size) == 1:
        size = (size[0], size[0])
737

vfdev's avatar
vfdev committed
738
739
740
    if len(size) != 2:
        raise ValueError("Please provide only two dimensions (h, w) for size.")

741
    _, image_height, image_width = get_dimensions(img)
742
743
744
745
746
    crop_height, crop_width = size
    if crop_width > image_width or crop_height > image_height:
        msg = "Requested crop size {} is bigger than input size {}"
        raise ValueError(msg.format(size, (image_height, image_width)))

vfdev's avatar
vfdev committed
747
748
749
750
751
752
753
754
    tl = crop(img, 0, 0, crop_height, crop_width)
    tr = crop(img, 0, image_width - crop_width, crop_height, crop_width)
    bl = crop(img, image_height - crop_height, 0, crop_height, crop_width)
    br = crop(img, image_height - crop_height, image_width - crop_width, crop_height, crop_width)

    center = center_crop(img, [crop_height, crop_width])

    return tl, tr, bl, br, center
755
756


vfdev's avatar
vfdev committed
757
758
759
def ten_crop(img: Tensor, size: List[int], vertical_flip: bool = False) -> List[Tensor]:
    """Generate ten cropped images from the given image.
    Crop the given image into four corners and the central crop plus the
760
    flipped version of these (horizontal flipping is used by default).
761
    If the image is torch Tensor, it is expected
vfdev's avatar
vfdev committed
762
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions
763
764
765
766
767

    .. Note::
        This transform returns a tuple of images and there may be a
        mismatch in the number of inputs and targets your ``Dataset`` returns.

768
    Args:
vfdev's avatar
vfdev committed
769
        img (PIL Image or Tensor): Image to be cropped.
770
        size (sequence or int): Desired output size of the crop. If size is an
771
            int instead of sequence like (h, w), a square crop (size, size) is
772
            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).
773
        vertical_flip (bool): Use vertical flipping instead of horizontal
774
775

    Returns:
776
        tuple: tuple (tl, tr, bl, br, center, tl_flip, tr_flip, bl_flip, br_flip, center_flip)
777
778
        Corresponding top left, top right, bottom left, bottom right and
        center crop and same for the flipped image.
779
    """
780
781
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(ten_crop)
782
783
    if isinstance(size, numbers.Number):
        size = (int(size), int(size))
vfdev's avatar
vfdev committed
784
785
786
787
788
    elif isinstance(size, (tuple, list)) and len(size) == 1:
        size = (size[0], size[0])

    if len(size) != 2:
        raise ValueError("Please provide only two dimensions (h, w) for size.")
789
790
791
792
793
794
795
796
797
798
799
800

    first_five = five_crop(img, size)

    if vertical_flip:
        img = vflip(img)
    else:
        img = hflip(img)

    second_five = five_crop(img, size)
    return first_five + second_five


801
def adjust_brightness(img: Tensor, brightness_factor: float) -> Tensor:
802
    """Adjust brightness of an image.
803
804

    Args:
vfdev's avatar
vfdev committed
805
        img (PIL Image or Tensor): Image to be adjusted.
806
807
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
            where ... means it can have an arbitrary number of leading dimensions.
808
809
810
811
812
        brightness_factor (float):  How much to adjust the brightness. Can be
            any non negative number. 0 gives a black image, 1 gives the
            original image while 2 increases the brightness by a factor of 2.

    Returns:
vfdev's avatar
vfdev committed
813
        PIL Image or Tensor: Brightness adjusted image.
814
    """
815
816
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(adjust_brightness)
817
818
    if not isinstance(img, torch.Tensor):
        return F_pil.adjust_brightness(img, brightness_factor)
819

820
    return F_t.adjust_brightness(img, brightness_factor)
821
822


823
def adjust_contrast(img: Tensor, contrast_factor: float) -> Tensor:
824
    """Adjust contrast of an image.
825
826

    Args:
vfdev's avatar
vfdev committed
827
        img (PIL Image or Tensor): Image to be adjusted.
828
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
829
            where ... means it can have an arbitrary number of leading dimensions.
830
831
832
833
834
        contrast_factor (float): How much to adjust the contrast. Can be any
            non negative number. 0 gives a solid gray image, 1 gives the
            original image while 2 increases the contrast by a factor of 2.

    Returns:
vfdev's avatar
vfdev committed
835
        PIL Image or Tensor: Contrast adjusted image.
836
    """
837
838
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(adjust_contrast)
839
840
    if not isinstance(img, torch.Tensor):
        return F_pil.adjust_contrast(img, contrast_factor)
841

842
    return F_t.adjust_contrast(img, contrast_factor)
843
844


845
def adjust_saturation(img: Tensor, saturation_factor: float) -> Tensor:
846
847
848
    """Adjust color saturation of an image.

    Args:
vfdev's avatar
vfdev committed
849
        img (PIL Image or Tensor): Image to be adjusted.
850
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
851
            where ... means it can have an arbitrary number of leading dimensions.
852
853
854
855
856
        saturation_factor (float):  How much to adjust the saturation. 0 will
            give a black and white image, 1 will give the original image while
            2 will enhance the saturation by a factor of 2.

    Returns:
vfdev's avatar
vfdev committed
857
        PIL Image or Tensor: Saturation adjusted image.
858
    """
859
860
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(adjust_saturation)
861
862
    if not isinstance(img, torch.Tensor):
        return F_pil.adjust_saturation(img, saturation_factor)
863

864
    return F_t.adjust_saturation(img, saturation_factor)
865
866


867
def adjust_hue(img: Tensor, hue_factor: float) -> Tensor:
868
869
870
871
872
873
874
875
876
    """Adjust hue of an image.

    The image hue is adjusted by converting the image to HSV and
    cyclically shifting the intensities in the hue channel (H).
    The image is then converted back to original image mode.

    `hue_factor` is the amount of shift in H channel and must be in the
    interval `[-0.5, 0.5]`.

877
878
879
    See `Hue`_ for more details.

    .. _Hue: https://en.wikipedia.org/wiki/Hue
880
881

    Args:
882
        img (PIL Image or Tensor): Image to be adjusted.
883
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
884
            where ... means it can have an arbitrary number of leading dimensions.
885
            If img is PIL Image mode "1", "I", "F" and modes with transparency (alpha channel) are not supported.
886
887
888
889
890
891
892
        hue_factor (float):  How much to shift the hue channel. Should be in
            [-0.5, 0.5]. 0.5 and -0.5 give complete reversal of hue channel in
            HSV space in positive and negative direction respectively.
            0 means no shift. Therefore, both -0.5 and 0.5 will give an image
            with complementary colors while 0 gives the original image.

    Returns:
893
        PIL Image or Tensor: Hue adjusted image.
894
    """
895
896
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(adjust_hue)
897
898
    if not isinstance(img, torch.Tensor):
        return F_pil.adjust_hue(img, hue_factor)
899

900
    return F_t.adjust_hue(img, hue_factor)
901
902


903
def adjust_gamma(img: Tensor, gamma: float, gain: float = 1) -> Tensor:
904
    r"""Perform gamma correction on an image.
905
906
907
908

    Also known as Power Law Transform. Intensities in RGB mode are adjusted
    based on the following equation:

909
910
911
912
    .. math::
        I_{\text{out}} = 255 \times \text{gain} \times \left(\frac{I_{\text{in}}}{255}\right)^{\gamma}

    See `Gamma Correction`_ for more details.
913

914
    .. _Gamma Correction: https://en.wikipedia.org/wiki/Gamma_correction
915
916

    Args:
917
        img (PIL Image or Tensor): PIL Image to be adjusted.
918
919
920
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
            where ... means it can have an arbitrary number of leading dimensions.
            If img is PIL Image, modes with transparency (alpha channel) are not supported.
921
922
923
        gamma (float): Non negative real number, same as :math:`\gamma` in the equation.
            gamma larger than 1 make the shadows darker,
            while gamma smaller than 1 make dark regions lighter.
924
        gain (float): The constant multiplier.
925
926
    Returns:
        PIL Image or Tensor: Gamma correction adjusted image.
927
    """
928
929
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(adjust_gamma)
930
931
    if not isinstance(img, torch.Tensor):
        return F_pil.adjust_gamma(img, gamma, gain)
932

933
    return F_t.adjust_gamma(img, gamma, gain)
934
935


vfdev's avatar
vfdev committed
936
def _get_inverse_affine_matrix(
937
938
939
940
941
    center: List[float],
    angle: float,
    translate: List[float],
    scale: float,
    shear: List[float],
vfdev's avatar
vfdev committed
942
) -> List[float]:
943
944
    # Helper method to compute inverse matrix for affine transformation

945
946
947
    # Pillow requires inverse affine transformation matrix:
    # Affine matrix is : M = T * C * RotateScaleShear * C^-1
    #
948
949
    # where T is translation matrix: [1, 0, tx | 0, 1, ty | 0, 0, 1]
    #       C is translation matrix to keep center: [1, 0, cx | 0, 1, cy | 0, 0, 1]
950
951
952
    #       RotateScaleShear is rotation with scale and shear matrix
    #
    #       RotateScaleShear(a, s, (sx, sy)) =
953
    #       = R(a) * S(s) * SHy(sy) * SHx(sx)
954
955
    #       = [ s*cos(a - sy)/cos(sy), s*(-cos(a - sy)*tan(sx)/cos(sy) - sin(a)), 0 ]
    #         [ s*sin(a + sy)/cos(sy), s*(-sin(a - sy)*tan(sx)/cos(sy) + cos(a)), 0 ]
956
957
958
959
960
    #         [ 0                    , 0                                      , 1 ]
    # where R is a rotation matrix, S is a scaling matrix, and SHx and SHy are the shears:
    # SHx(s) = [1, -tan(s)] and SHy(s) = [1      , 0]
    #          [0, 1      ]              [-tan(s), 1]
    #
961
    # Thus, the inverse is M^-1 = C * RotateScaleShear^-1 * C^-1 * T^-1
962

963
    rot = math.radians(angle)
964
965
    sx = math.radians(shear[0])
    sy = math.radians(shear[1])
966
967
968
969
970

    cx, cy = center
    tx, ty = translate

    # RSS without scaling
vfdev's avatar
vfdev committed
971
972
973
974
    a = math.cos(rot - sy) / math.cos(sy)
    b = -math.cos(rot - sy) * math.tan(sx) / math.cos(sy) - math.sin(rot)
    c = math.sin(rot - sy) / math.cos(sy)
    d = -math.sin(rot - sy) * math.tan(sx) / math.cos(sy) + math.cos(rot)
975
976

    # Inverted rotation matrix with scale and shear
977
    # det([[a, b], [c, d]]) == 1, since det(rotation) = 1 and det(shear) = 1
vfdev's avatar
vfdev committed
978
979
    matrix = [d, -b, 0.0, -c, a, 0.0]
    matrix = [x / scale for x in matrix]
980
981

    # Apply inverse of translation and of center translation: RSS^-1 * C^-1 * T^-1
vfdev's avatar
vfdev committed
982
983
    matrix[2] += matrix[0] * (-cx - tx) + matrix[1] * (-cy - ty)
    matrix[5] += matrix[3] * (-cx - tx) + matrix[4] * (-cy - ty)
984
985

    # Apply center translation: C * RSS^-1 * C^-1 * T^-1
vfdev's avatar
vfdev committed
986
987
    matrix[2] += cx
    matrix[5] += cy
988

vfdev's avatar
vfdev committed
989
    return matrix
990

vfdev's avatar
vfdev committed
991

vfdev's avatar
vfdev committed
992
def rotate(
993
994
995
996
997
998
999
    img: Tensor,
    angle: float,
    interpolation: InterpolationMode = InterpolationMode.NEAREST,
    expand: bool = False,
    center: Optional[List[int]] = None,
    fill: Optional[List[float]] = None,
    resample: Optional[int] = None,
vfdev's avatar
vfdev committed
1000
1001
) -> Tensor:
    """Rotate the image by angle.
1002
    If the image is torch Tensor, it is expected
vfdev's avatar
vfdev committed
1003
1004
1005
1006
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.

    Args:
        img (PIL Image or Tensor): image to be rotated.
1007
        angle (number): rotation angle value in degrees, counter-clockwise.
1008
1009
1010
        interpolation (InterpolationMode): Desired interpolation enum defined by
            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.
            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
1011
            For backward compatibility integer values (e.g. ``PIL.Image.NEAREST``) are still acceptable.
vfdev's avatar
vfdev committed
1012
1013
1014
1015
        expand (bool, optional): Optional expansion flag.
            If true, expands the output image to make it large enough to hold the entire rotated image.
            If false or omitted, make the output image the same size as the input image.
            Note that the expand flag assumes rotation around the center and no translation.
1016
        center (sequence, optional): Optional center of rotation. Origin is the upper left corner.
vfdev's avatar
vfdev committed
1017
            Default is the center of the image.
1018
1019
        fill (sequence or number, optional): Pixel fill value for the area outside the transformed
            image. If given a number, the value is used for all bands respectively.
1020
1021
1022
1023

            .. note::
                In torchscript mode single int/float value is not supported, please use a sequence
                of length 1: ``[value, ]``.
1024
1025
1026
1027
        resample (int, optional):
            .. warning::
                This parameter was deprecated in ``0.12`` and will be removed in ``0.14``. Please use ``interpolation``
                instead.
vfdev's avatar
vfdev committed
1028
1029
1030
1031
1032
1033
1034

    Returns:
        PIL Image or Tensor: Rotated image.

    .. _filters: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#filters

    """
1035
1036
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(rotate)
1037
1038
    if resample is not None:
        warnings.warn(
1039
1040
            "The parameter 'resample' is deprecated since 0.12 and will be removed 0.14. "
            "Please use 'interpolation' instead."
1041
1042
1043
1044
1045
1046
        )
        interpolation = _interpolation_modes_from_int(resample)

    # Backward compatibility with integer value
    if isinstance(interpolation, int):
        warnings.warn(
1047
1048
            "Argument interpolation should be of type InterpolationMode instead of int. "
            "Please, use InterpolationMode enum."
1049
1050
1051
        )
        interpolation = _interpolation_modes_from_int(interpolation)

vfdev's avatar
vfdev committed
1052
1053
1054
1055
1056
1057
    if not isinstance(angle, (int, float)):
        raise TypeError("Argument angle should be int or float")

    if center is not None and not isinstance(center, (list, tuple)):
        raise TypeError("Argument center should be a sequence")

1058
1059
    if not isinstance(interpolation, InterpolationMode):
        raise TypeError("Argument interpolation should be a InterpolationMode")
1060

vfdev's avatar
vfdev committed
1061
    if not isinstance(img, torch.Tensor):
1062
1063
        pil_interpolation = pil_modes_mapping[interpolation]
        return F_pil.rotate(img, angle=angle, interpolation=pil_interpolation, expand=expand, center=center, fill=fill)
vfdev's avatar
vfdev committed
1064
1065
1066

    center_f = [0.0, 0.0]
    if center is not None:
1067
        _, height, width = get_dimensions(img)
1068
        # Center values should be in pixel coordinates but translated such that (0, 0) corresponds to image center.
1069
        center_f = [1.0 * (c - s * 0.5) for c, s in zip(center, [width, height])]
1070

vfdev's avatar
vfdev committed
1071
1072
1073
    # due to current incoherence of rotation angle direction between affine and rotate implementations
    # we need to set -angle.
    matrix = _get_inverse_affine_matrix(center_f, -angle, [0.0, 0.0], 1.0, [0.0, 0.0])
1074
    return F_t.rotate(img, matrix=matrix, interpolation=interpolation.value, expand=expand, fill=fill)
vfdev's avatar
vfdev committed
1075
1076


vfdev's avatar
vfdev committed
1077
def affine(
1078
1079
1080
1081
1082
1083
1084
1085
1086
    img: Tensor,
    angle: float,
    translate: List[int],
    scale: float,
    shear: List[float],
    interpolation: InterpolationMode = InterpolationMode.NEAREST,
    fill: Optional[List[float]] = None,
    resample: Optional[int] = None,
    fillcolor: Optional[List[float]] = None,
1087
    center: Optional[List[int]] = None,
vfdev's avatar
vfdev committed
1088
1089
) -> Tensor:
    """Apply affine transformation on the image keeping image center invariant.
1090
    If the image is torch Tensor, it is expected
vfdev's avatar
vfdev committed
1091
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
1092
1093

    Args:
vfdev's avatar
vfdev committed
1094
        img (PIL Image or Tensor): image to transform.
1095
1096
        angle (number): rotation angle in degrees between -180 and 180, clockwise direction.
        translate (sequence of integers): horizontal and vertical translations (post-rotation translation)
1097
        scale (float): overall scale
1098
1099
        shear (float or sequence): shear angle value in degrees between -180 to 180, clockwise direction.
            If a sequence is specified, the first value corresponds to a shear parallel to the x axis, while
vfdev's avatar
vfdev committed
1100
            the second value corresponds to a shear parallel to the y axis.
1101
1102
1103
        interpolation (InterpolationMode): Desired interpolation enum defined by
            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.
            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
1104
            For backward compatibility integer values (e.g. ``PIL.Image.NEAREST``) are still acceptable.
1105
1106
        fill (sequence or number, optional): Pixel fill value for the area outside the transformed
            image. If given a number, the value is used for all bands respectively.
1107
1108
1109
1110

            .. note::
                In torchscript mode single int/float value is not supported, please use a sequence
                of length 1: ``[value, ]``.
1111
1112
1113
1114
1115
1116
1117
        fillcolor (sequence or number, optional):
            .. warning::
                This parameter was deprecated in ``0.12`` and will be removed in ``0.14``. Please use ``fill`` instead.
        resample (int, optional):
            .. warning::
                This parameter was deprecated in ``0.12`` and will be removed in ``0.14``. Please use ``interpolation``
                instead.
1118
1119
        center (sequence, optional): Optional center of rotation. Origin is the upper left corner.
            Default is the center of the image.
vfdev's avatar
vfdev committed
1120
1121
1122

    Returns:
        PIL Image or Tensor: Transformed image.
1123
    """
1124
1125
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(affine)
1126
1127
    if resample is not None:
        warnings.warn(
1128
1129
            "The parameter 'resample' is deprecated since 0.12 and will be removed in 0.14. "
            "Please use 'interpolation' instead."
1130
1131
1132
1133
1134
1135
        )
        interpolation = _interpolation_modes_from_int(resample)

    # Backward compatibility with integer value
    if isinstance(interpolation, int):
        warnings.warn(
1136
1137
            "Argument interpolation should be of type InterpolationMode instead of int. "
            "Please, use InterpolationMode enum."
1138
1139
1140
1141
        )
        interpolation = _interpolation_modes_from_int(interpolation)

    if fillcolor is not None:
1142
1143
1144
1145
        warnings.warn(
            "The parameter 'fillcolor' is deprecated since 0.12 and will be removed in 0.14. "
            "Please use 'fill' instead."
        )
1146
1147
        fill = fillcolor

vfdev's avatar
vfdev committed
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
    if not isinstance(angle, (int, float)):
        raise TypeError("Argument angle should be int or float")

    if not isinstance(translate, (list, tuple)):
        raise TypeError("Argument translate should be a sequence")

    if len(translate) != 2:
        raise ValueError("Argument translate should be a sequence of length 2")

    if scale <= 0.0:
        raise ValueError("Argument scale should be positive")

    if not isinstance(shear, (numbers.Number, (list, tuple))):
        raise TypeError("Shear should be either a single value or a sequence of two values")

1163
1164
    if not isinstance(interpolation, InterpolationMode):
        raise TypeError("Argument interpolation should be a InterpolationMode")
1165

vfdev's avatar
vfdev committed
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
    if isinstance(angle, int):
        angle = float(angle)

    if isinstance(translate, tuple):
        translate = list(translate)

    if isinstance(shear, numbers.Number):
        shear = [shear, 0.0]

    if isinstance(shear, tuple):
        shear = list(shear)

    if len(shear) == 1:
        shear = [shear[0], shear[0]]

    if len(shear) != 2:
1182
        raise ValueError(f"Shear should be a sequence containing two values. Got {shear}")
vfdev's avatar
vfdev committed
1183

1184
1185
1186
    if center is not None and not isinstance(center, (list, tuple)):
        raise TypeError("Argument center should be a sequence")

1187
    _, height, width = get_dimensions(img)
vfdev's avatar
vfdev committed
1188
    if not isinstance(img, torch.Tensor):
1189
        # center = (width * 0.5 + 0.5, height * 0.5 + 0.5)
vfdev's avatar
vfdev committed
1190
1191
        # it is visually better to estimate the center without 0.5 offset
        # otherwise image rotated by 90 degrees is shifted vs output image of torch.rot90 or F_t.affine
1192
        if center is None:
1193
            center = [width * 0.5, height * 0.5]
vfdev's avatar
vfdev committed
1194
        matrix = _get_inverse_affine_matrix(center, angle, translate, scale, shear)
1195
1196
        pil_interpolation = pil_modes_mapping[interpolation]
        return F_pil.affine(img, matrix=matrix, interpolation=pil_interpolation, fill=fill)
1197

1198
1199
    center_f = [0.0, 0.0]
    if center is not None:
1200
        _, height, width = get_dimensions(img)
1201
        # Center values should be in pixel coordinates but translated such that (0, 0) corresponds to image center.
1202
        center_f = [1.0 * (c - s * 0.5) for c, s in zip(center, [width, height])]
1203

1204
    translate_f = [1.0 * t for t in translate]
1205
    matrix = _get_inverse_affine_matrix(center_f, angle, translate_f, scale, shear)
1206
    return F_t.affine(img, matrix=matrix, interpolation=interpolation.value, fill=fill)
1207
1208


1209
@torch.jit.unused
1210
def to_grayscale(img, num_output_channels=1):
1211
    """Convert PIL image of any mode (RGB, HSV, LAB, etc) to grayscale version of image.
1212
    This transform does not support torch Tensor.
1213
1214

    Args:
1215
        img (PIL Image): PIL Image to be converted to grayscale.
1216
        num_output_channels (int): number of channels of the output image. Value can be 1 or 3. Default is 1.
1217
1218

    Returns:
1219
1220
        PIL Image: Grayscale version of the image.

1221
1222
        - if num_output_channels = 1 : returned image is single channel
        - if num_output_channels = 3 : returned image is 3 channel with r = g = b
1223
    """
1224
1225
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(to_grayscale)
1226
1227
    if isinstance(img, Image.Image):
        return F_pil.to_grayscale(img, num_output_channels)
1228

1229
1230
1231
1232
1233
    raise TypeError("Input should be PIL Image")


def rgb_to_grayscale(img: Tensor, num_output_channels: int = 1) -> Tensor:
    """Convert RGB image to grayscale version of image.
1234
1235
    If the image is torch Tensor, it is expected
    to have [..., 3, H, W] shape, where ... means an arbitrary number of leading dimensions
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247

    Note:
        Please, note that this method supports only RGB images as input. For inputs in other color spaces,
        please, consider using meth:`~torchvision.transforms.functional.to_grayscale` with PIL Image.

    Args:
        img (PIL Image or Tensor): RGB Image to be converted to grayscale.
        num_output_channels (int): number of channels of the output image. Value can be 1 or 3. Default, 1.

    Returns:
        PIL Image or Tensor: Grayscale version of the image.

1248
1249
        - if num_output_channels = 1 : returned image is single channel
        - if num_output_channels = 3 : returned image is 3 channel with r = g = b
1250
    """
1251
1252
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(rgb_to_grayscale)
1253
1254
1255
1256
    if not isinstance(img, torch.Tensor):
        return F_pil.to_grayscale(img, num_output_channels)

    return F_t.rgb_to_grayscale(img, num_output_channels)
1257
1258


1259
def erase(img: Tensor, i: int, j: int, h: int, w: int, v: Tensor, inplace: bool = False) -> Tensor:
1260
    """Erase the input Tensor Image with given value.
1261
    This transform does not support PIL Image.
1262
1263
1264
1265
1266
1267
1268
1269

    Args:
        img (Tensor Image): Tensor image of size (C, H, W) to be erased
        i (int): i in (i,j) i.e coordinates of the upper left corner.
        j (int): j in (i,j) i.e coordinates of the upper left corner.
        h (int): Height of the erased region.
        w (int): Width of the erased region.
        v: Erasing value.
Zhun Zhong's avatar
Zhun Zhong committed
1270
        inplace(bool, optional): For in-place operations. By default is set False.
1271
1272
1273
1274

    Returns:
        Tensor Image: Erased image.
    """
1275
1276
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(erase)
1277
    if not isinstance(img, torch.Tensor):
1278
        raise TypeError(f"img should be Tensor Image. Got {type(img)}")
1279

1280
    return F_t.erase(img, i, j, h, w, v, inplace=inplace)
1281
1282
1283


def gaussian_blur(img: Tensor, kernel_size: List[int], sigma: Optional[List[float]] = None) -> Tensor:
1284
1285
1286
    """Performs Gaussian blurring on the image by given kernel.
    If the image is torch Tensor, it is expected
    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
1287
1288
1289
1290
1291

    Args:
        img (PIL Image or Tensor): Image to be blurred
        kernel_size (sequence of ints or int): Gaussian kernel size. Can be a sequence of integers
            like ``(kx, ky)`` or a single integer for square kernels.
1292
1293
1294
1295

            .. note::
                In torchscript mode kernel_size as single int is not supported, use a sequence of
                length 1: ``[ksize, ]``.
1296
1297
1298
1299
        sigma (sequence of floats or float, optional): Gaussian kernel standard deviation. Can be a
            sequence of floats like ``(sigma_x, sigma_y)`` or a single float to define the
            same sigma in both X/Y directions. If None, then it is computed using
            ``kernel_size`` as ``sigma = 0.3 * ((kernel_size - 1) * 0.5 - 1) + 0.8``.
1300
1301
1302
1303
1304
            Default, None.

            .. note::
                In torchscript mode sigma as single float is
                not supported, use a sequence of length 1: ``[sigma, ]``.
1305
1306
1307
1308

    Returns:
        PIL Image or Tensor: Gaussian Blurred version of the image.
    """
1309
1310
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(gaussian_blur)
1311
    if not isinstance(kernel_size, (int, list, tuple)):
1312
        raise TypeError(f"kernel_size should be int or a sequence of integers. Got {type(kernel_size)}")
1313
1314
1315
    if isinstance(kernel_size, int):
        kernel_size = [kernel_size, kernel_size]
    if len(kernel_size) != 2:
1316
        raise ValueError(f"If kernel_size is a sequence its length should be 2. Got {len(kernel_size)}")
1317
1318
    for ksize in kernel_size:
        if ksize % 2 == 0 or ksize < 0:
1319
            raise ValueError(f"kernel_size should have odd and positive integers. Got {kernel_size}")
1320
1321
1322
1323
1324

    if sigma is None:
        sigma = [ksize * 0.15 + 0.35 for ksize in kernel_size]

    if sigma is not None and not isinstance(sigma, (int, float, list, tuple)):
1325
        raise TypeError(f"sigma should be either float or sequence of floats. Got {type(sigma)}")
1326
1327
1328
1329
1330
    if isinstance(sigma, (int, float)):
        sigma = [float(sigma), float(sigma)]
    if isinstance(sigma, (list, tuple)) and len(sigma) == 1:
        sigma = [sigma[0], sigma[0]]
    if len(sigma) != 2:
1331
        raise ValueError(f"If sigma is a sequence, its length should be 2. Got {len(sigma)}")
1332
    for s in sigma:
1333
        if s <= 0.0:
1334
            raise ValueError(f"sigma should have positive values. Got {sigma}")
1335
1336
1337
1338

    t_img = img
    if not isinstance(img, torch.Tensor):
        if not F_pil._is_pil_image(img):
1339
            raise TypeError(f"img should be PIL Image or Tensor. Got {type(img)}")
1340
1341
1342
1343
1344
1345
1346
1347

        t_img = to_tensor(img)

    output = F_t.gaussian_blur(t_img, kernel_size, sigma)

    if not isinstance(img, torch.Tensor):
        output = to_pil_image(output)
    return output
1348
1349
1350


def invert(img: Tensor) -> Tensor:
1351
    """Invert the colors of an RGB/grayscale image.
1352
1353
1354

    Args:
        img (PIL Image or Tensor): Image to have its colors inverted.
1355
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
1356
1357
            where ... means it can have an arbitrary number of leading dimensions.
            If img is PIL Image, it is expected to be in mode "L" or "RGB".
1358
1359
1360
1361

    Returns:
        PIL Image or Tensor: Color inverted image.
    """
1362
1363
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(invert)
1364
1365
1366
1367
1368
1369
1370
    if not isinstance(img, torch.Tensor):
        return F_pil.invert(img)

    return F_t.invert(img)


def posterize(img: Tensor, bits: int) -> Tensor:
1371
    """Posterize an image by reducing the number of bits for each color channel.
1372
1373
1374

    Args:
        img (PIL Image or Tensor): Image to have its colors posterized.
1375
            If img is torch Tensor, it should be of type torch.uint8 and
1376
1377
1378
            it is expected to be in [..., 1 or 3, H, W] format, where ... means
            it can have an arbitrary number of leading dimensions.
            If img is PIL Image, it is expected to be in mode "L" or "RGB".
1379
1380
1381
1382
        bits (int): The number of bits to keep for each channel (0-8).
    Returns:
        PIL Image or Tensor: Posterized image.
    """
1383
1384
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(posterize)
1385
    if not (0 <= bits <= 8):
1386
        raise ValueError(f"The number if bits should be between 0 and 8. Got {bits}")
1387
1388
1389
1390
1391
1392
1393
1394

    if not isinstance(img, torch.Tensor):
        return F_pil.posterize(img, bits)

    return F_t.posterize(img, bits)


def solarize(img: Tensor, threshold: float) -> Tensor:
1395
    """Solarize an RGB/grayscale image by inverting all pixel values above a threshold.
1396
1397
1398

    Args:
        img (PIL Image or Tensor): Image to have its colors inverted.
1399
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
1400
1401
            where ... means it can have an arbitrary number of leading dimensions.
            If img is PIL Image, it is expected to be in mode "L" or "RGB".
1402
1403
1404
1405
        threshold (float): All pixels equal or above this value are inverted.
    Returns:
        PIL Image or Tensor: Solarized image.
    """
1406
1407
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(solarize)
1408
1409
1410
1411
1412
1413
1414
    if not isinstance(img, torch.Tensor):
        return F_pil.solarize(img, threshold)

    return F_t.solarize(img, threshold)


def adjust_sharpness(img: Tensor, sharpness_factor: float) -> Tensor:
1415
    """Adjust the sharpness of an image.
1416
1417
1418

    Args:
        img (PIL Image or Tensor): Image to be adjusted.
1419
1420
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
            where ... means it can have an arbitrary number of leading dimensions.
1421
1422
1423
1424
1425
1426
1427
        sharpness_factor (float):  How much to adjust the sharpness. Can be
            any non negative number. 0 gives a blurred image, 1 gives the
            original image while 2 increases the sharpness by a factor of 2.

    Returns:
        PIL Image or Tensor: Sharpness adjusted image.
    """
1428
1429
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(adjust_sharpness)
1430
1431
1432
1433
1434
1435
1436
    if not isinstance(img, torch.Tensor):
        return F_pil.adjust_sharpness(img, sharpness_factor)

    return F_t.adjust_sharpness(img, sharpness_factor)


def autocontrast(img: Tensor) -> Tensor:
1437
    """Maximize contrast of an image by remapping its
1438
1439
1440
1441
1442
    pixels per channel so that the lowest becomes black and the lightest
    becomes white.

    Args:
        img (PIL Image or Tensor): Image on which autocontrast is applied.
1443
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
1444
1445
            where ... means it can have an arbitrary number of leading dimensions.
            If img is PIL Image, it is expected to be in mode "L" or "RGB".
1446
1447
1448
1449

    Returns:
        PIL Image or Tensor: An image that was autocontrasted.
    """
1450
1451
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(autocontrast)
1452
1453
1454
1455
1456
1457
1458
    if not isinstance(img, torch.Tensor):
        return F_pil.autocontrast(img)

    return F_t.autocontrast(img)


def equalize(img: Tensor) -> Tensor:
1459
    """Equalize the histogram of an image by applying
1460
1461
1462
1463
1464
    a non-linear mapping to the input in order to create a uniform
    distribution of grayscale values in the output.

    Args:
        img (PIL Image or Tensor): Image on which equalize is applied.
1465
            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,
1466
            where ... means it can have an arbitrary number of leading dimensions.
1467
            The tensor dtype must be ``torch.uint8`` and values are expected to be in ``[0, 255]``.
1468
            If img is PIL Image, it is expected to be in mode "P", "L" or "RGB".
1469
1470
1471
1472

    Returns:
        PIL Image or Tensor: An image that was equalized.
    """
1473
1474
    if not torch.jit.is_scripting() and not torch.jit.is_tracing():
        _log_api_usage_once(equalize)
1475
1476
1477
1478
    if not isinstance(img, torch.Tensor):
        return F_pil.equalize(img)

    return F_t.equalize(img)