diffusers 0.27.1__py3-none-any.whl → 0.28.0__py3-none-any.whl

diffusers/pipelines/t2i_adapter/pipeline_stable_diffusion_xl_adapter.py

@@ -140,6 +140,7 @@ def retrieve_timesteps(
     num_inference_steps: Optional[int] = None,
     device: Optional[Union[str, torch.device]] = None,
     timesteps: Optional[List[int]] = None,
+    sigmas: Optional[List[float]] = None,
     **kwargs,
 ):
     """
@@ -150,19 +151,23 @@ def retrieve_timesteps(
         scheduler (`SchedulerMixin`):
             The scheduler to get timesteps from.
         num_inference_steps (`int`):
-            The number of diffusion steps used when generating samples with a pre-trained model. If used,
-            `timesteps` must be `None`.
+            The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps`
+            must be `None`.
         device (`str` or `torch.device`, *optional*):
             The device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
         timesteps (`List[int]`, *optional*):
-                Custom timesteps used to support arbitrary spacing between timesteps. If `None`, then the default
-                timestep spacing strategy of the scheduler is used. If `timesteps` is passed, `num_inference_steps`
-                must be `None`.
+            Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed,
+            `num_inference_steps` and `sigmas` must be `None`.
+        sigmas (`List[float]`, *optional*):
+            Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed,
+            `num_inference_steps` and `timesteps` must be `None`.
 
     Returns:
         `Tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the
         second element is the number of inference steps.
     """
+    if timesteps is not None and sigmas is not None:
+        raise ValueError("Only one of `timesteps` or `sigmas` can be passed. Please choose one to set custom values")
     if timesteps is not None:
         accepts_timesteps = "timesteps" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
         if not accepts_timesteps:
@@ -173,6 +178,16 @@ def retrieve_timesteps(
         scheduler.set_timesteps(timesteps=timesteps, device=device, **kwargs)
         timesteps = scheduler.timesteps
         num_inference_steps = len(timesteps)
+    elif sigmas is not None:
+        accept_sigmas = "sigmas" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+        if not accept_sigmas:
+            raise ValueError(
+                f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+                f" sigmas schedules. Please check whether you are using the correct scheduler."
+            )
+        scheduler.set_timesteps(sigmas=sigmas, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+        num_inference_steps = len(timesteps)
     else:
         scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
         timesteps = scheduler.timesteps
@@ -281,10 +296,10 @@ class StableDiffusionXLAdapterPipeline(
         do_classifier_free_guidance: bool = True,
         negative_prompt: Optional[str] = None,
         negative_prompt_2: Optional[str] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
-        pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        pooled_prompt_embeds: Optional[torch.Tensor] = None,
+        negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
         lora_scale: Optional[float] = None,
         clip_skip: Optional[int] = None,
     ):
@@ -310,17 +325,17 @@ class StableDiffusionXLAdapterPipeline(
             negative_prompt_2 (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
                 `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
-            pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+            pooled_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
                 If not provided, pooled text embeddings will be generated from `prompt` input argument.
-            negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
                 input argument.
@@ -700,7 +715,12 @@ class StableDiffusionXLAdapterPipeline(
 
     # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_latents
     def prepare_latents(self, batch_size, num_channels_latents, height, width, dtype, device, generator, latents=None):
-        shape = (batch_size, num_channels_latents, height // self.vae_scale_factor, width // self.vae_scale_factor)
+        shape = (
+            batch_size,
+            num_channels_latents,
+            int(height) // self.vae_scale_factor,
+            int(width) // self.vae_scale_factor,
+        )
         if isinstance(generator, list) and len(generator) != batch_size:
             raise ValueError(
                 f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
@@ -784,20 +804,22 @@ class StableDiffusionXLAdapterPipeline(
         return height, width
 
     # Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
-    def get_guidance_scale_embedding(self, w, embedding_dim=512, dtype=torch.float32):
+    def get_guidance_scale_embedding(
+        self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32
+    ) -> torch.Tensor:
         """
         See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
 
         Args:
-            timesteps (`torch.Tensor`):
-                generate embedding vectors at these timesteps
+            w (`torch.Tensor`):
+                Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings.
             embedding_dim (`int`, *optional*, defaults to 512):
-                dimension of the embeddings to generate
-            dtype:
-                data type of the generated embeddings
+                Dimension of the embeddings to generate.
+            dtype (`torch.dtype`, *optional*, defaults to `torch.float32`):
+                Data type of the generated embeddings.
 
         Returns:
-            `torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)`
+            `torch.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`.
         """
         assert len(w.shape) == 1
         w = w * 1000.0
@@ -834,6 +856,7 @@ class StableDiffusionXLAdapterPipeline(
         width: Optional[int] = None,
         num_inference_steps: int = 50,
         timesteps: List[int] = None,
+        sigmas: List[float] = None,
         denoising_end: Optional[float] = None,
         guidance_scale: float = 5.0,
         negative_prompt: Optional[Union[str, List[str]]] = None,
@@ -841,16 +864,16 @@ class StableDiffusionXLAdapterPipeline(
         num_images_per_prompt: Optional[int] = 1,
         eta: float = 0.0,
         generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
-        latents: Optional[torch.FloatTensor] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
-        pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
+        latents: Optional[torch.Tensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
+        pooled_prompt_embeds: Optional[torch.Tensor] = None,
+        negative_pooled_prompt_embeds: Optional[torch.Tensor] = None,
         ip_adapter_image: Optional[PipelineImageInput] = None,
-        ip_adapter_image_embeds: Optional[List[torch.FloatTensor]] = None,
+        ip_adapter_image_embeds: Optional[List[torch.Tensor]] = None,
         output_type: Optional[str] = "pil",
         return_dict: bool = True,
-        callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
         callback_steps: int = 1,
         cross_attention_kwargs: Optional[Dict[str, Any]] = None,
         guidance_rescale: float = 0.0,
@@ -874,9 +897,9 @@ class StableDiffusionXLAdapterPipeline(
             prompt_2 (`str` or `List[str]`, *optional*):
                 The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
                 used in both text-encoders
-            image (`torch.FloatTensor`, `PIL.Image.Image`, `List[torch.FloatTensor]` or `List[PIL.Image.Image]` or `List[List[PIL.Image.Image]]`):
+            image (`torch.Tensor`, `PIL.Image.Image`, `List[torch.Tensor]` or `List[PIL.Image.Image]` or `List[List[PIL.Image.Image]]`):
                 The Adapter input condition. Adapter uses this input condition to generate guidance to Unet. If the
-                type is specified as `Torch.FloatTensor`, it is passed to Adapter as is. PIL.Image.Image` can also be
+                type is specified as `torch.Tensor`, it is passed to Adapter as is. PIL.Image.Image` can also be
                 accepted as an image. The control image is automatically resized to fit the output image.
             height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
                 The height in pixels of the generated image. Anything below 512 pixels won't work well for
@@ -893,6 +916,10 @@ class StableDiffusionXLAdapterPipeline(
                 Custom timesteps to use for the denoising process with schedulers which support a `timesteps` argument
                 in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is
                 passed will be used. Must be in descending order.
+            sigmas (`List[float]`, *optional*):
+                Custom sigmas to use for the denoising process with schedulers which support a `sigmas` argument in
+                their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is passed
+                will be used.
             denoising_end (`float`, *optional*):
                 When specified, determines the fraction (between 0.0 and 1.0) of the total denoising process to be
                 completed before it is intentionally prematurely terminated. As a result, the returned sample will
@@ -921,30 +948,30 @@ class StableDiffusionXLAdapterPipeline(
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
                 to make generation deterministic.
-            latents (`torch.FloatTensor`, *optional*):
+            latents (`torch.Tensor`, *optional*):
                 Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
                 generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
                 tensor will ge generated by sampling using the supplied random `generator`.
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
-            pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+            pooled_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
                 If not provided, pooled text embeddings will be generated from `prompt` input argument.
-            negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_pooled_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
                 input argument.
             ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters.
-            ip_adapter_image_embeds (`List[torch.FloatTensor]`, *optional*):
-                Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters.
-                Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding
-                if `do_classifier_free_guidance` is set to `True`.
-                If not provided, embeddings are computed from the `ip_adapter_image` input argument.
+            ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*):
+                Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of
+                IP-adapters. Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should
+                contain the negative image embedding if `do_classifier_free_guidance` is set to `True`. If not
+                provided, embeddings are computed from the `ip_adapter_image` input argument.
             output_type (`str`, *optional*, defaults to `"pil"`):
                 The output format of the generate image. Choose between
                 [PIL](https://pillow.readthedocs.io/en/stable/): `PIL.Image.Image` or `np.array`.
@@ -953,7 +980,7 @@ class StableDiffusionXLAdapterPipeline(
                 instead of a plain tuple.
             callback (`Callable`, *optional*):
                 A function that will be called every `callback_steps` steps during inference. The function will be
-                called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`.
+                called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
             callback_steps (`int`, *optional*, defaults to 1):
                 The frequency at which the `callback` function will be called. If not specified, the callback will be
                 called at every step.
@@ -1094,7 +1121,9 @@ class StableDiffusionXLAdapterPipeline(
             )
 
         # 4. Prepare timesteps
-        timesteps, num_inference_steps = retrieve_timesteps(self.scheduler, num_inference_steps, device, timesteps)
+        timesteps, num_inference_steps = retrieve_timesteps(
+            self.scheduler, num_inference_steps, device, timesteps, sigmas
+        )
 
         # 5. Prepare latent variables
         num_channels_latents = self.unet.config.in_channels

diffusers/pipelines/text_to_video_synthesis/pipeline_output.py

@@ -15,9 +15,10 @@ class TextToVideoSDPipelineOutput(BaseOutput):
     """
      Output class for text-to-video pipelines.
 
-     Args:
+    Args:
          frames (`torch.Tensor`, `np.ndarray`, or List[List[PIL.Image.Image]]):
-             List of video outputs - It can be a nested list of length `batch_size,` with each sub-list containing denoised
+             List of video outputs - It can be a nested list of length `batch_size,` with each sub-list containing
+             denoised
      PIL image sequences of length `num_frames.` It can also be a NumPy array or Torch tensor of shape
     `(batch_size, num_frames, channels, height, width)`
     """

diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_synth.py

@@ -15,11 +15,9 @@
 import inspect
 from typing import Any, Callable, Dict, List, Optional, Union
 
-import numpy as np
 import torch
 from transformers import CLIPTextModel, CLIPTokenizer
 
-from ...image_processor import VaeImageProcessor
 from ...loaders import LoraLoaderMixin, TextualInversionLoaderMixin
 from ...models import AutoencoderKL, UNet3DConditionModel
 from ...models.lora import adjust_lora_scale_text_encoder
@@ -33,6 +31,7 @@ from ...utils import (
     unscale_lora_layers,
 )
 from ...utils.torch_utils import randn_tensor
+from ...video_processor import VideoProcessor
 from ..pipeline_utils import DiffusionPipeline, StableDiffusionMixin
 from . import TextToVideoSDPipelineOutput
 
@@ -59,28 +58,6 @@ EXAMPLE_DOC_STRING = """
 """
 
 
-# Copied from diffusers.pipelines.animatediff.pipeline_animatediff.tensor2vid
-def tensor2vid(video: torch.Tensor, processor: "VaeImageProcessor", output_type: str = "np"):
-    batch_size, channels, num_frames, height, width = video.shape
-    outputs = []
-    for batch_idx in range(batch_size):
-        batch_vid = video[batch_idx].permute(1, 0, 2, 3)
-        batch_output = processor.postprocess(batch_vid, output_type)
-
-        outputs.append(batch_output)
-
-    if output_type == "np":
-        outputs = np.stack(outputs)
-
-    elif output_type == "pt":
-        outputs = torch.stack(outputs)
-
-    elif not output_type == "pil":
-        raise ValueError(f"{output_type} does not exist. Please choose one of ['np', 'pt', 'pil']")
-
-    return outputs
-
-
 class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInversionLoaderMixin, LoraLoaderMixin):
     r"""
     Pipeline for text-to-video generation.
@@ -127,7 +104,7 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
             scheduler=scheduler,
         )
         self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
-        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self.video_processor = VideoProcessor(do_resize=False, vae_scale_factor=self.vae_scale_factor)
 
     # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline._encode_prompt
     def _encode_prompt(
@@ -137,8 +114,8 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
         num_images_per_prompt,
         do_classifier_free_guidance,
         negative_prompt=None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         lora_scale: Optional[float] = None,
         **kwargs,
     ):
@@ -170,8 +147,8 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
         num_images_per_prompt,
         do_classifier_free_guidance,
         negative_prompt=None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         lora_scale: Optional[float] = None,
         clip_skip: Optional[int] = None,
     ):
@@ -191,10 +168,10 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                 less than `1`).
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
@@ -465,12 +442,12 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
         negative_prompt: Optional[Union[str, List[str]]] = None,
         eta: float = 0.0,
         generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
-        latents: Optional[torch.FloatTensor] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        latents: Optional[torch.Tensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         output_type: Optional[str] = "np",
         return_dict: bool = True,
-        callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
         callback_steps: int = 1,
         cross_attention_kwargs: Optional[Dict[str, Any]] = None,
         clip_skip: Optional[int] = None,
@@ -505,25 +482,25 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
                 generation deterministic.
-            latents (`torch.FloatTensor`, *optional*):
+            latents (`torch.Tensor`, *optional*):
                 Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for video
                 generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
                 tensor is generated by sampling using the supplied random `generator`. Latents should be of shape
                 `(batch_size, num_channel, num_frames, height, width)`.
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
                 provided, text embeddings are generated from the `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
                 not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
             output_type (`str`, *optional*, defaults to `"np"`):
-                The output format of the generated video. Choose between `torch.FloatTensor` or `np.array`.
+                The output format of the generated video. Choose between `torch.Tensor` or `np.array`.
             return_dict (`bool`, *optional*, defaults to `True`):
                 Whether or not to return a [`~pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput`] instead
                 of a plain tuple.
             callback (`Callable`, *optional*):
                 A function that calls every `callback_steps` steps during inference. The function is called with the
-                following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`.
+                following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
             callback_steps (`int`, *optional*, defaults to 1):
                 The frequency at which the `callback` function is called. If not specified, the callback is called at
                 every step.
@@ -652,7 +629,7 @@ class TextToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInve
             video = latents
         else:
             video_tensor = self.decode_latents(latents)
-            video = tensor2vid(video_tensor, self.image_processor, output_type)
+            video = self.video_processor.postprocess_video(video=video_tensor, output_type=output_type)
 
         # 9. Offload all models
         self.maybe_free_model_hooks()

diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_synth_img2img.py

@@ -16,11 +16,9 @@ import inspect
 from typing import Any, Callable, Dict, List, Optional, Union
 
 import numpy as np
-import PIL.Image
 import torch
 from transformers import CLIPTextModel, CLIPTokenizer
 
-from ...image_processor import VaeImageProcessor
 from ...loaders import LoraLoaderMixin, TextualInversionLoaderMixin
 from ...models import AutoencoderKL, UNet3DConditionModel
 from ...models.lora import adjust_lora_scale_text_encoder
@@ -34,6 +32,7 @@ from ...utils import (
     unscale_lora_layers,
 )
 from ...utils.torch_utils import randn_tensor
+from ...video_processor import VideoProcessor
 from ..pipeline_utils import DiffusionPipeline, StableDiffusionMixin
 from . import TextToVideoSDPipelineOutput
 
@@ -94,69 +93,6 @@ def retrieve_latents(
         raise AttributeError("Could not access latents of provided encoder_output")
 
 
-# Copied from diffusers.pipelines.animatediff.pipeline_animatediff.tensor2vid
-def tensor2vid(video: torch.Tensor, processor: "VaeImageProcessor", output_type: str = "np"):
-    batch_size, channels, num_frames, height, width = video.shape
-    outputs = []
-    for batch_idx in range(batch_size):
-        batch_vid = video[batch_idx].permute(1, 0, 2, 3)
-        batch_output = processor.postprocess(batch_vid, output_type)
-
-        outputs.append(batch_output)
-
-    if output_type == "np":
-        outputs = np.stack(outputs)
-
-    elif output_type == "pt":
-        outputs = torch.stack(outputs)
-
-    elif not output_type == "pil":
-        raise ValueError(f"{output_type} does not exist. Please choose one of ['np', 'pt', 'pil']")
-
-    return outputs
-
-
-def preprocess_video(video):
-    supported_formats = (np.ndarray, torch.Tensor, PIL.Image.Image)
-
-    if isinstance(video, supported_formats):
-        video = [video]
-    elif not (isinstance(video, list) and all(isinstance(i, supported_formats) for i in video)):
-        raise ValueError(
-            f"Input is in incorrect format: {[type(i) for i in video]}. Currently, we only support {', '.join(supported_formats)}"
-        )
-
-    if isinstance(video[0], PIL.Image.Image):
-        video = [np.array(frame) for frame in video]
-
-    if isinstance(video[0], np.ndarray):
-        video = np.concatenate(video, axis=0) if video[0].ndim == 5 else np.stack(video, axis=0)
-
-        if video.dtype == np.uint8:
-            video = np.array(video).astype(np.float32) / 255.0
-
-        if video.ndim == 4:
-            video = video[None, ...]
-
-        video = torch.from_numpy(video.transpose(0, 4, 1, 2, 3))
-
-    elif isinstance(video[0], torch.Tensor):
-        video = torch.cat(video, axis=0) if video[0].ndim == 5 else torch.stack(video, axis=0)
-
-        # don't need any preprocess if the video is latents
-        channel = video.shape[1]
-        if channel == 4:
-            return video
-
-        # move channels before num_frames
-        video = video.permute(0, 2, 1, 3, 4)
-
-    # normalize video
-    video = 2.0 * video - 1.0
-
-    return video
-
-
 class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInversionLoaderMixin, LoraLoaderMixin):
     r"""
     Pipeline for text-guided video-to-video generation.
@@ -203,7 +139,7 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
             scheduler=scheduler,
         )
         self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
-        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+        self.video_processor = VideoProcessor(do_resize=False, vae_scale_factor=self.vae_scale_factor)
 
     # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline._encode_prompt
     def _encode_prompt(
@@ -213,8 +149,8 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
         num_images_per_prompt,
         do_classifier_free_guidance,
         negative_prompt=None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         lora_scale: Optional[float] = None,
         **kwargs,
     ):
@@ -246,8 +182,8 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
         num_images_per_prompt,
         do_classifier_free_guidance,
         negative_prompt=None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         lora_scale: Optional[float] = None,
         clip_skip: Optional[int] = None,
     ):
@@ -267,10 +203,10 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                 less than `1`).
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                 provided, text embeddings will be generated from `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                 weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                 argument.
@@ -563,19 +499,19 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
     def __call__(
         self,
         prompt: Union[str, List[str]] = None,
-        video: Union[List[np.ndarray], torch.FloatTensor] = None,
+        video: Union[List[np.ndarray], torch.Tensor] = None,
         strength: float = 0.6,
         num_inference_steps: int = 50,
         guidance_scale: float = 15.0,
         negative_prompt: Optional[Union[str, List[str]]] = None,
         eta: float = 0.0,
         generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
-        latents: Optional[torch.FloatTensor] = None,
-        prompt_embeds: Optional[torch.FloatTensor] = None,
-        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
+        latents: Optional[torch.Tensor] = None,
+        prompt_embeds: Optional[torch.Tensor] = None,
+        negative_prompt_embeds: Optional[torch.Tensor] = None,
         output_type: Optional[str] = "np",
         return_dict: bool = True,
-        callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
+        callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
         callback_steps: int = 1,
         cross_attention_kwargs: Optional[Dict[str, Any]] = None,
         clip_skip: Optional[int] = None,
@@ -586,7 +522,7 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
         Args:
             prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
-            video (`List[np.ndarray]` or `torch.FloatTensor`):
+            video (`List[np.ndarray]` or `torch.Tensor`):
                 `video` frames or tensor representing a video batch to be used as the starting point for the process.
                 Can also accept video latents as `image`, if passing latents directly, it will not be encoded again.
             strength (`float`, *optional*, defaults to 0.8):
@@ -610,25 +546,25 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
                 generation deterministic.
-            latents (`torch.FloatTensor`, *optional*):
+            latents (`torch.Tensor`, *optional*):
                 Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for video
                 generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
                 tensor is generated by sampling using the supplied random `generator`. Latents should be of shape
                 `(batch_size, num_channel, num_frames, height, width)`.
-            prompt_embeds (`torch.FloatTensor`, *optional*):
+            prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
                 provided, text embeddings are generated from the `prompt` input argument.
-            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+            negative_prompt_embeds (`torch.Tensor`, *optional*):
                 Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
                 not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
             output_type (`str`, *optional*, defaults to `"np"`):
-                The output format of the generated video. Choose between `torch.FloatTensor` or `np.array`.
+                The output format of the generated video. Choose between `torch.Tensor` or `np.array`.
             return_dict (`bool`, *optional*, defaults to `True`):
                 Whether or not to return a [`~pipelines.text_to_video_synthesis.TextToVideoSDPipelineOutput`] instead
                 of a plain tuple.
             callback (`Callable`, *optional*):
                 A function that calls every `callback_steps` steps during inference. The function is called with the
-                following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`.
+                following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`.
             callback_steps (`int`, *optional*, defaults to 1):
                 The frequency at which the `callback` function is called. If not specified, the callback is called at
                 every step.
@@ -687,7 +623,7 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
             prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])
 
         # 4. Preprocess video
-        video = preprocess_video(video)
+        video = self.video_processor.preprocess_video(video)
 
         # 5. Prepare timesteps
         self.scheduler.set_timesteps(num_inference_steps, device=device)
@@ -749,7 +685,7 @@ class VideoToVideoSDPipeline(DiffusionPipeline, StableDiffusionMixin, TextualInv
             video = latents
         else:
             video_tensor = self.decode_latents(latents)
-            video = tensor2vid(video_tensor, self.image_processor, output_type)
+            video = self.video_processor.postprocess_video(video=video_tensor, output_type=output_type)
 
         # 10. Offload all models
         self.maybe_free_model_hooks()