[Pipeline] Add TextToVideoZeroPipeline (#2954)

* add TextToVideoZeroPipeline and CrossFrameAttnProcessor

* add docs for text-to-video zero

* add teaser image for text-to-video zero docs

* Fix review changes. Add Documentation. Add test

* clean up the codes in pipeline_text_to_video.py. Add descriptive comments and docstrings

* make style && make quality

* make fix-copies

* make requested changes to docs. use huggingface server links for resources, delete res folder

* make style && make quality && make fix-copies

* make style && make quality

* Apply suggestions from code review

---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

docs/source/en/_toctree.yml

@@ -206,6 +206,8 @@
       title: Stochastic Karras VE
     - local: api/pipelines/text_to_video
       title: Text-to-Video
+    - local: api/pipelines/text_to_video_zero
+      title: Text-to-Video Zero
     - local: api/pipelines/unclip
       title: UnCLIP
     - local: api/pipelines/latent_diffusion_uncond

docs/source/en/api/pipelines/overview.mdx

@@ -83,6 +83,7 @@ available a colab notebook to directly try them out.
 | [versatile_diffusion](./versatile_diffusion) | [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) | Image Variations Generation | 
 | [versatile_diffusion](./versatile_diffusion) | [Versatile Diffusion: Text, Images and Variations All in One Diffusion Model](https://arxiv.org/abs/2211.08332) | Dual Image and Text Guided Generation | 
 | [vq_diffusion](./vq_diffusion) | [Vector Quantized Diffusion Model for Text-to-Image Synthesis](https://arxiv.org/abs/2111.14822) | Text-to-Image Generation | 
+| [text_to_video_zero](./text_to_video_zero) | [Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generators](https://arxiv.org/abs/2303.13439) | Text-to-Video Generation |
 
 
 **Note**: Pipelines are simple examples of how to play around with the diffusion systems as described in the corresponding papers. 

docs/source/en/api/pipelines/text_to_video_zero.mdx

+<!--Copyright 2023 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+-->
+
+# Zero-Shot Text-to-Video Generation 
+
+## Overview
+
+
+[Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generators](https://arxiv.org/abs/2303.13439) by
+Levon Khachatryan,
+Andranik Movsisyan,
+Vahram Tadevosyan,
+Roberto Henschel,
+[Zhangyang Wang](https://www.ece.utexas.edu/people/faculty/atlas-wang), Shant Navasardyan, [Humphrey Shi](https://www.humphreyshi.com).
+
+Our method Text2Video-Zero enables zero-shot video generation using either
+1. A textual prompt, or
+2. A prompt combined with guidance from poses or edges, or 
+3. Video Instruct-Pix2Pix, i.e., instruction-guided video editing.
+
+Results are temporally consistent and follow closely the guidance and textual prompts.
+
+![teaser-img](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/t2v_zero_teaser.png)
+
+The abstract of the paper is the following:
+
+*Recent text-to-video generation approaches rely on computationally heavy training and require large-scale video datasets. In this paper, we introduce a new task of zero-shot text-to-video generation and propose a low-cost approach (without any training or optimization) by leveraging the power of existing text-to-image synthesis methods (e.g., Stable Diffusion), making them suitable for the video domain.
+Our key modifications include (i) enriching the latent codes of the generated frames with motion dynamics to keep the global scene and the background time consistent; and (ii) reprogramming frame-level self-attention using a new cross-frame attention of each frame on the first frame, to preserve the context, appearance, and identity of the foreground object.
+Experiments show that this leads to low overhead, yet high-quality and remarkably consistent video generation. Moreover, our approach is not limited to text-to-video synthesis but is also applicable to other tasks such as conditional and content-specialized video generation, and Video Instruct-Pix2Pix, i.e., instruction-guided video editing.
+As experiments show, our method performs comparably or sometimes better than recent approaches, despite not being trained on additional video data.*
+
+
+
+Resources:
+
+* [Project Page](https://text2video-zero.github.io/)
+* [Paper](https://arxiv.org/abs/2303.13439)
+* [Original Code](https://github.com/Picsart-AI-Research/Text2Video-Zero)
+
+
+## Available Pipelines:
+
+| Pipeline | Tasks | Demo
+|---|---|:---:|
+| [TextToVideoZeroPipeline](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_zero.py) | *Zero-shot Text-to-Video Generation* | [🤗 Space](https://huggingface.co/spaces/PAIR/Text2Video-Zero)
+
+
+## Usage example
+
+### Text-To-Video
+
+To generate a video from prompt, run the following python command
+```python
+import torch
+from diffusers import TextToVideoZeroPipeline
+
+model_id = "runwayml/stable-diffusion-v1-5"
+pipe = TextToVideoZeroPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")
+
+prompt = "A panda is playing guitar on times square"
+result = pipe(prompt=prompt).images
+imageio.mimsave("video.mp4", result, fps=4)
+```
+You can change these parameters in the pipeline call:
+* Motion field strength (see the [paper](https://arxiv.org/abs/2303.13439), Sect. 3.3.1):
+    * `motion_field_strength_x` and `motion_field_strength_y`. Default: `motion_field_strength_x=12`, `motion_field_strength_y=12`
+* `T` and `T'` (see the [paper](https://arxiv.org/abs/2303.13439), Sect. 3.3.1)
+    * `t0` and `t1` in the range `{0, ..., num_inference_steps}`. Default: `t0=45`, `t1=48`
+* Video length:
+    * `video_length`, the number of frames video_length to be generated. Default: `video_length=8`
+
+
+### Text-To-Video with Pose Control
+To generate a video from prompt with additional pose control
+
+1. Download a demo video
+
+    ```python
+    from huggingface_hub import hf_hub_download
+
+    filename = "__assets__/poses_skeleton_gifs/dance1_corr.mp4"
+    repo_id = "PAIR/Text2Video-Zero"
+    video_path = hf_hub_download(repo_type="space", repo_id=repo_id, filename=filename)
+    ```
+
+
+2. Read video containing extracted pose images
+    ```python
+    import imageio
+
+    reader = imageio.get_reader(video_path, "ffmpeg")
+    frame_count = 8
+    pose_images = [Image.fromarray(reader.get_data(i)) for i in range(frame_count)]
+    ```
+    To extract pose from actual video, read [ControlNet documentation](./stable_diffusion/controlnet).
+
+3. Run `StableDiffusionControlNetPipeline` with our custom attention processor
+
+    ```python
+    import torch
+    from diffusers import StableDiffusionControlNetPipeline, ControlNetModel
+    from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero import CrossFrameAttnProcessor
+
+    model_id = "runwayml/stable-diffusion-v1-5"
+    controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-openpose", torch_dtype=torch.float16)
+    pipe = StableDiffusionControlNetPipeline.from_pretrained(
+        model_id, controlnet=controlnet, torch_dtype=torch.float16
+    ).to("cuda")
+
+    # Set the attention processor
+    pipe.unet.set_attn_processor(CrossFrameAttnProcessor(batch_size=2))
+    pipe.controlnet.set_attn_processor(CrossFrameAttnProcessor(batch_size=2))
+
+    # fix latents for all frames
+    latents = torch.randn((1, 4, 64, 64), device="cuda", dtype=torch.float16).repeat(len(pose_images), 1, 1, 1)
+
+    prompt = "Darth Vader dancing in a desert"
+    result = pipe(prompt=[prompt] * len(pose_images), image=pose_images, latents=latents).images
+    imageio.mimsave("video.mp4", result, fps=4)
+    ```
+
+
+### Text-To-Video with Edge Control
+
+To generate a video from prompt with additional pose control,
+follow the steps described above for pose-guided generation using [Canny edge ControlNet model](https://huggingface.co/lllyasviel/sd-controlnet-canny).
+
+
+### Video Instruct-Pix2Pix
+
+To perform text-guided video editing (with [InstructPix2Pix](./stable_diffusion/pix2pix)):
+
+1. Download a demo video
+
+    ```python
+    from huggingface_hub import hf_hub_download
+
+    filename = "__assets__/pix2pix video/camel.mp4"
+    repo_id = "PAIR/Text2Video-Zero"
+    video_path = hf_hub_download(repo_type="space", repo_id=repo_id, filename=filename)
+    ```
+
+2. Read video from path
+    ```python
+    import imageio
+
+    reader = imageio.get_reader(video_path, "ffmpeg")
+    frame_count = 8
+    video = [Image.fromarray(reader.get_data(i)) for i in range(frame_count)]
+    ```
+
+3. Run `StableDiffusionInstructPix2PixPipeline` with our custom attention processor
+    ```python
+    import torch
+    from diffusers import StableDiffusionInstructPix2PixPipeline
+    from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero import CrossFrameAttnProcessor
+
+    model_id = "timbrooks/instruct-pix2pix"
+    pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")
+    pipe.unet.set_attn_processor(CrossFrameAttnProcessor(batch_size=3))
+
+    prompt = "make it Van Gogh Starry Night style"
+    result = pipe(prompt=[prompt] * len(video), image=video).images
+    imageio.mimsave("edited_video.mp4", result, fps=4)
+    ```
+
+
+### Dreambooth specialization 
+
+Methods **Text-To-Video**, **Text-To-Video with Pose Control** and **Text-To-Video with Edge Control**
+can run with custom [DreamBooth](../training/dreambooth) models, as shown below for
+[Canny edge ControlNet model](https://huggingface.co/lllyasviel/sd-controlnet-canny) and
+[Avatar style DreamBooth](https://huggingface.co/PAIR/text2video-zero-controlnet-canny-avatar) model
+
+1. Download demo video from huggingface
+
+    ```python
+    from huggingface_hub import hf_hub_download
+
+    filename = "__assets__/canny_videos_mp4/girl_turning.mp4"
+    repo_id = "PAIR/Text2Video-Zero"
+    video_path = hf_hub_download(repo_type="space", repo_id=repo_id, filename=filename)
+    ```
+
+2. Read video from path
+    ```python
+    import imageio
+
+    reader = imageio.get_reader(video_path, "ffmpeg")
+    frame_count = 8
+    video = [Image.fromarray(reader.get_data(i)) for i in range(frame_count)]
+    ```
+
+3. Run `StableDiffusionControlNetPipeline` with custom trained DreamBooth model
+    ```python
+    import torch
+    from diffusers import StableDiffusionControlNetPipeline, ControlNetModel
+    from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero import CrossFrameAttnProcessor
+
+    # set model id to custom model
+    model_id = "PAIR/text2video-zero-controlnet-canny-avatar"
+    controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16)
+    pipe = StableDiffusionControlNetPipeline.from_pretrained(
+        model_id, controlnet=controlnet, torch_dtype=torch.float16
+    ).to("cuda")
+
+    # Set the attention processor
+    pipe.unet.set_attn_processor(CrossFrameAttnProcessor(batch_size=2))
+    pipe.controlnet.set_attn_processor(CrossFrameAttnProcessor(batch_size=2))
+
+    # fix latents for all frames
+    latents = torch.randn((1, 4, 64, 64), device="cuda", dtype=torch.float16).repeat(len(pose_images), 1, 1, 1)
+
+    prompt = "oil painting of a beautiful girl avatar style"
+    result = pipe(prompt=[prompt] * len(pose_images), image=pose_images, latents=latents).images
+    imageio.mimsave("video.mp4", result, fps=4)
+    ```
+
+You can filter out some available DreamBooth-trained models with [this link](https://huggingface.co/models?search=dreambooth).
+
+
+
+## TextToVideoZeroPipeline
+[[autodoc]] TextToVideoZeroPipeline
+	- all
+	- __call__
\ No newline at end of file

src/diffusers/__init__.py

@@ -137,6 +137,7 @@ else:
         StableUnCLIPImg2ImgPipeline,
         StableUnCLIPPipeline,
         TextToVideoSDPipeline,
+        TextToVideoZeroPipeline,
         UnCLIPImageVariationPipeline,
         UnCLIPPipeline,
         VersatileDiffusionDualGuidedPipeline,

src/diffusers/pipelines/__init__.py

@@ -68,7 +68,7 @@ else:
         StableUnCLIPPipeline,
     )
     from .stable_diffusion_safe import StableDiffusionPipelineSafe
-    from .text_to_video_synthesis import TextToVideoSDPipeline
+    from .text_to_video_synthesis import TextToVideoSDPipeline, TextToVideoZeroPipeline
     from .unclip import UnCLIPImageVariationPipeline, UnCLIPPipeline
     from .versatile_diffusion import (
         VersatileDiffusionDualGuidedPipeline,

src/diffusers/pipelines/text_to_video_synthesis/__init__.py

@@ -29,3 +29,4 @@ except OptionalDependencyNotAvailable:
     from ...utils.dummy_torch_and_transformers_objects import *  # noqa F403
 else:
     from .pipeline_text_to_video_synth import TextToVideoSDPipeline  # noqa: F401
+    from .pipeline_text_to_video_zero import TextToVideoZeroPipeline

src/diffusers/utils/dummy_torch_and_transformers_objects.py

@@ -407,6 +407,21 @@ class TextToVideoSDPipeline(metaclass=DummyObject):
         requires_backends(cls, ["torch", "transformers"])
 
 
+class TextToVideoZeroPipeline(metaclass=DummyObject):
+    _backends = ["torch", "transformers"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch", "transformers"])
+
+    @classmethod
+    def from_config(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+    @classmethod
+    def from_pretrained(cls, *args, **kwargs):
+        requires_backends(cls, ["torch", "transformers"])
+
+
 class UnCLIPImageVariationPipeline(metaclass=DummyObject):
     _backends = ["torch", "transformers"]
 

tests/pipelines/text_to_video/test_text_to_video_zero.py

+# coding=utf-8
+# Copyright 2023 HuggingFace Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import unittest
+
+import torch
+
+from diffusers import DDIMScheduler, TextToVideoZeroPipeline
+from diffusers.utils import require_torch_gpu, slow
+
+from ...test_pipelines_common import assert_mean_pixel_difference
+
+
+@slow
+@require_torch_gpu
+class TextToVideoZeroPipelineSlowTests(unittest.TestCase):
+    def test_full_model(self):
+        model_id = "runwayml/stable-diffusion-v1-5"
+        pipe = TextToVideoZeroPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")
+        pipe.scheduler = DDIMScheduler.from_config(pipe.scheduler.config)
+        generator = torch.Generator(device="cuda").manual_seed(0)
+
+        prompt = "A bear is playing a guitar on Times Square"
+        result = pipe(prompt=prompt, generator=generator).images
+
+        expected_result = torch.load(
+            "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/tree/main/text-to-video/A bear is playing a guitar on Times Square.pt"
+        )
+
+        assert_mean_pixel_difference(result, expected_result)