Hugging Face's logo

Diffusers documentation

ConsisID

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v0.32.2).
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

ConsisID

Identity-Preserving Text-to-Video Generation by Frequency Decomposition from Peking University & University of Rochester & etc, by Shenghai Yuan, Jinfa Huang, Xianyi He, Yunyang Ge, Yujun Shi, Liuhan Chen, Jiebo Luo, Li Yuan.

The abstract from the paper is:

Identity-preserving text-to-video (IPT2V) generation aims to create high-fidelity videos with consistent human identity. It is an important task in video generation but remains an open problem for generative models. This paper pushes the technical frontier of IPT2V in two directions that have not been resolved in the literature: (1) A tuning-free pipeline without tedious case-by-case finetuning, and (2) A frequency-aware heuristic identity-preserving Diffusion Transformer (DiT)-based control scheme. To achieve these goals, we propose ConsisID, a tuning-free DiT-based controllable IPT2V model to keep human-identity consistent in the generated video. Inspired by prior findings in frequency analysis of vision/diffusion transformers, it employs identity-control signals in the frequency domain, where facial features can be decomposed into low-frequency global features (e.g., profile, proportions) and high-frequency intrinsic features (e.g., identity markers that remain unaffected by pose changes). First, from a low-frequency perspective, we introduce a global facial extractor, which encodes the reference image and facial key points into a latent space, generating features enriched with low-frequency information. These features are then integrated into the shallow layers of the network to alleviate training challenges associated with DiT. Second, from a high-frequency perspective, we design a local facial extractor to capture high-frequency details and inject them into the transformer blocks, enhancing the model’s ability to preserve fine-grained features. To leverage the frequency information for identity preservation, we propose a hierarchical training strategy, transforming a vanilla pre-trained video generation model into an IPT2V model. Extensive experiments demonstrate that our frequency-aware heuristic scheme provides an optimal control solution for DiT-based models. Thanks to this scheme, our ConsisID achieves excellent results in generating high-quality, identity-preserving videos, making strides towards more effective IPT2V. The model weight of ConsID is publicly available at https://github.com/PKU-YuanGroup/ConsisID.

Make sure to check out the Schedulers guide to learn how to explore the tradeoff between scheduler speed and quality, and see the reuse components across pipelines section to learn how to efficiently load the same components into multiple pipelines.

This pipeline was contributed by SHYuanBest. The original codebase can be found here. The original weights can be found under hf.co/BestWishYsh.

There are two official ConsisID checkpoints for identity-preserving text-to-video.

checkpoints recommended inference dtype
BestWishYsh/ConsisID-preview torch.bfloat16
BestWishYsh/ConsisID-1.5 torch.bfloat16

Memory optimization

ConsisID requires about 44 GB of GPU memory to decode 49 frames (6 seconds of video at 8 FPS) with output resolution 720x480 (W x H), which makes it not possible to run on consumer GPUs or free-tier T4 Colab. The following memory optimizations could be used to reduce the memory footprint. For replication, you can refer to this script.

Feature (overlay the previous) Max Memory Allocated Max Memory Reserved
- 37 GB 44 GB
enable_model_cpu_offload 22 GB 25 GB
enable_sequential_cpu_offload 16 GB 22 GB
vae.enable_slicing 16 GB 22 GB
vae.enable_tiling 5 GB 7 GB

ConsisIDPipeline

class diffusers.ConsisIDPipeline

< >

( tokenizer: T5Tokenizer text_encoder: T5EncoderModel vae: AutoencoderKLCogVideoX transformer: ConsisIDTransformer3DModel scheduler: CogVideoXDPMScheduler )

Parameters

  • vae (AutoencoderKL) — Variational Auto-Encoder (VAE) Model to encode and decode videos to and from latent representations.
  • text_encoder (T5EncoderModel) — Frozen text-encoder. ConsisID uses T5; specifically the t5-v1_1-xxl variant.
  • tokenizer (T5Tokenizer) — Tokenizer of class T5Tokenizer.
  • transformer (ConsisIDTransformer3DModel) — A text conditioned ConsisIDTransformer3DModel to denoise the encoded video latents.
  • scheduler (SchedulerMixin) — A scheduler to be used in combination with transformer to denoise the encoded video latents.

Pipeline for image-to-video generation using ConsisID.

This model inherits from DiffusionPipeline. Check the superclass documentation for the generic methods the library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)

__call__

< >

( image: typing.Union[PIL.Image.Image, numpy.ndarray, torch.Tensor, typing.List[PIL.Image.Image], typing.List[numpy.ndarray], typing.List[torch.Tensor]] prompt: typing.Union[str, typing.List[str], NoneType] = None negative_prompt: typing.Union[str, typing.List[str], NoneType] = None height: int = 480 width: int = 720 num_frames: int = 49 num_inference_steps: int = 50 guidance_scale: float = 6.0 use_dynamic_cfg: bool = False num_videos_per_prompt: int = 1 eta: float = 0.0 generator: typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None latents: typing.Optional[torch.FloatTensor] = None prompt_embeds: typing.Optional[torch.FloatTensor] = None negative_prompt_embeds: typing.Optional[torch.FloatTensor] = None output_type: str = 'pil' return_dict: bool = True attention_kwargs: typing.Optional[typing.Dict[str, typing.Any]] = None callback_on_step_end: typing.Union[typing.Callable[[int, int, typing.Dict], NoneType], diffusers.callbacks.PipelineCallback, diffusers.callbacks.MultiPipelineCallbacks, NoneType] = None callback_on_step_end_tensor_inputs: typing.List[str] = ['latents'] max_sequence_length: int = 226 id_vit_hidden: typing.Optional[torch.Tensor] = None id_cond: typing.Optional[torch.Tensor] = None kps_cond: typing.Optional[torch.Tensor] = None ) → ConsisIDPipelineOutput or tuple

Parameters

  • image (PipelineImageInput) — The input image to condition the generation on. Must be an image, a list of images or a torch.Tensor.
  • prompt (str or List[str], optional) — The prompt or prompts to guide the image generation. If not defined, one has to pass prompt_embeds. instead.
  • negative_prompt (str or List[str], optional) — 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).
  • height (int, optional, defaults to self.transformer.config.sample_height * self.vae_scale_factor_spatial) — The height in pixels of the generated image. This is set to 480 by default for the best results.
  • width (int, optional, defaults to self.transformer.config.sample_height * self.vae_scale_factor_spatial) — The width in pixels of the generated image. This is set to 720 by default for the best results.
  • num_frames (int, defaults to 49) — Number of frames to generate. Must be divisible by self.vae_scale_factor_temporal. Generated video will contain 1 extra frame because ConsisID is conditioned with (num_seconds * fps + 1) frames where num_seconds is 6 and fps is 4. However, since videos can be saved at any fps, the only condition that needs to be satisfied is that of divisibility mentioned above.
  • num_inference_steps (int, optional, defaults to 50) — The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference.
  • guidance_scale (float, optional, defaults to 6) — Guidance scale as defined in Classifier-Free Diffusion Guidance. guidance_scale is defined as w of equation 2. of Imagen Paper. Guidance scale is enabled by setting guidance_scale > 1. Higher guidance scale encourages to generate images that are closely linked to the text prompt, usually at the expense of lower image quality.
  • use_dynamic_cfg (bool, optional, defaults to False) — If True, dynamically adjusts the guidance scale during inference. This allows the model to use a progressive guidance scale, improving the balance between text-guided generation and image quality over the course of the inference steps. Typically, early inference steps use a higher guidance scale for more faithful image generation, while later steps reduce it for more diverse and natural results.
  • num_videos_per_prompt (int, optional, defaults to 1) — The number of videos to generate per prompt.
  • generator (torch.Generator or List[torch.Generator], optional) — One or a list of torch generator(s) to make generation deterministic.
  • latents (torch.FloatTensor, 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) — 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) — 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.
  • output_type (str, optional, defaults to "pil") — The output format of the generate image. Choose between PIL: PIL.Image.Image or np.array.
  • return_dict (bool, optional, defaults to True) — Whether or not to return a ~pipelines.stable_diffusion_xl.StableDiffusionXLPipelineOutput instead of a plain tuple.
  • attention_kwargs (dict, optional) — A kwargs dictionary that if specified is passed along to the AttentionProcessor as defined under self.processor in diffusers.models.attention_processor.
  • callback_on_step_end (Callable, optional) — A function that calls at the end of each denoising steps during the inference. The function is called with the following arguments: callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict). callback_kwargs will include a list of all tensors as specified by callback_on_step_end_tensor_inputs.
  • callback_on_step_end_tensor_inputs (List, optional) — The list of tensor inputs for the callback_on_step_end function. The tensors specified in the list will be passed as callback_kwargs argument. You will only be able to include variables listed in the ._callback_tensor_inputs attribute of your pipeline class.
  • max_sequence_length (int, defaults to 226) — Maximum sequence length in encoded prompt. Must be consistent with self.transformer.config.max_text_seq_length otherwise may lead to poor results.
  • id_vit_hidden (Optional[torch.Tensor], optional) — The tensor representing the hidden features extracted from the face model, which are used to condition the local facial extractor. This is crucial for the model to obtain high-frequency information of the face. If not provided, the local facial extractor will not run normally.
  • id_cond (Optional[torch.Tensor], optional) — The tensor representing the hidden features extracted from the clip model, which are used to condition the local facial extractor. This is crucial for the model to edit facial features If not provided, the local facial extractor will not run normally.
  • kps_cond (Optional[torch.Tensor], optional) — A tensor that determines whether the global facial extractor use keypoint information for conditioning. If provided, this tensor controls whether facial keypoints such as eyes, nose, and mouth landmarks are used during the generation process. This helps ensure the model retains more facial low-frequency information.

Returns

ConsisIDPipelineOutput or tuple

ConsisIDPipelineOutput if return_dict is True, otherwise a tuple. When returning a tuple, the first element is a list with the generated images.

Function invoked when calling the pipeline for generation.

Examples:

>>> import torch
>>> from diffusers import ConsisIDPipeline
>>> from diffusers.pipelines.consisid.consisid_utils import prepare_face_models, process_face_embeddings_infer
>>> from diffusers.utils import export_to_video
>>> from huggingface_hub import snapshot_download

>>> snapshot_download(repo_id="BestWishYsh/ConsisID-preview", local_dir="BestWishYsh/ConsisID-preview")
>>> face_helper_1, face_helper_2, face_clip_model, face_main_model, eva_transform_mean, eva_transform_std = (
...     prepare_face_models("BestWishYsh/ConsisID-preview", device="cuda", dtype=torch.bfloat16)
... )
>>> pipe = ConsisIDPipeline.from_pretrained("BestWishYsh/ConsisID-preview", torch_dtype=torch.bfloat16)
>>> pipe.to("cuda")

>>> # ConsisID works well with long and well-described prompts. Make sure the face in the image is clearly visible (e.g., preferably half-body or full-body).
>>> prompt = "The video captures a boy walking along a city street, filmed in black and white on a classic 35mm camera. His expression is thoughtful, his brow slightly furrowed as if he's lost in contemplation. The film grain adds a textured, timeless quality to the image, evoking a sense of nostalgia. Around him, the cityscape is filled with vintage buildings, cobblestone sidewalks, and softly blurred figures passing by, their outlines faint and indistinct. Streetlights cast a gentle glow, while shadows play across the boy's path, adding depth to the scene. The lighting highlights the boy's subtle smile, hinting at a fleeting moment of curiosity. The overall cinematic atmosphere, complete with classic film still aesthetics and dramatic contrasts, gives the scene an evocative and introspective feel."
>>> image = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/consisid/consisid_input.png?download=true"

>>> id_cond, id_vit_hidden, image, face_kps = process_face_embeddings_infer(
...     face_helper_1,
...     face_clip_model,
...     face_helper_2,
...     eva_transform_mean,
...     eva_transform_std,
...     face_main_model,
...     "cuda",
...     torch.bfloat16,
...     image,
...     is_align_face=True,
... )

>>> video = pipe(
...     image=image,
...     prompt=prompt,
...     num_inference_steps=50,
...     guidance_scale=6.0,
...     use_dynamic_cfg=False,
...     id_vit_hidden=id_vit_hidden,
...     id_cond=id_cond,
...     kps_cond=face_kps,
...     generator=torch.Generator("cuda").manual_seed(42),
... )
>>> export_to_video(video.frames[0], "output.mp4", fps=8)

encode_prompt

< >

( prompt: typing.Union[str, typing.List[str]] negative_prompt: typing.Union[str, typing.List[str], NoneType] = None do_classifier_free_guidance: bool = True num_videos_per_prompt: int = 1 prompt_embeds: typing.Optional[torch.Tensor] = None negative_prompt_embeds: typing.Optional[torch.Tensor] = None max_sequence_length: int = 226 device: typing.Optional[torch.device] = None dtype: typing.Optional[torch.dtype] = None )

Parameters

  • prompt (str or List[str], optional) — prompt to be encoded
  • negative_prompt (str or List[str], optional) — 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).
  • do_classifier_free_guidance (bool, optional, defaults to True) — Whether to use classifier free guidance or not.
  • num_videos_per_prompt (int, optional, defaults to 1) — Number of videos that should be generated per prompt. torch device to place the resulting embeddings on
  • 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.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.
  • device — (torch.device, optional): torch device
  • dtype — (torch.dtype, optional): torch dtype

Encodes the prompt into text encoder hidden states.

ConsisIDPipelineOutput

class diffusers.pipelines.consisid.pipeline_output.ConsisIDPipelineOutput

< >

( frames: Tensor )

Parameters

  • 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 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).

Output class for ConsisID pipelines.

< > Update on GitHub

←CogView3 Consistency Models→