MusePose: a Pose-Driven Image-to-Video Framework for Virtual Human Generation.
Zhengyan Tong, Chao Li, Zhaokang Chen, Bin Wu†, Wenjiang Zhou (†Corresponding Author, [email protected])
Lyra Lab, Tencent Music Entertainment
github huggingface space (comming soon) Project (comming soon) Technical report (comming soon)
MusePose is an image-to-video generation framework for virtual human under control signal such as pose. The current released model was an implementation of AnimateAnyone by optimizing Moore-AnimateAnyone.
MusePose is the last building block of the Muse opensource serie. Together with MuseV and MuseTalk, we hope the community can join us and march towards the vision where a virtual human can be generated end2end with native ability of full body movement and interaction. Please stay tuned for our next milestone!
We really appreciate AnimateAnyone for their academic paper and Moore-AnimateAnyone for their code base, which have significantly expedited the development of the AIGC community and MusePose.
Update:
- We support Comfyui-MusePose now!
Join Lyra Lab, Tencent Music Entertainment!
We are currently seeking AIGC researchers including Internships, New Grads, and Senior (实习、校招、社招).
Please find details in the following two links or contact [email protected]
- AI Researcher (https://join.tencentmusic.com/social/post-details/?id=13488, https://join.tencentmusic.com/social/post-details/?id=13502)
MusePose is a diffusion-based and pose-guided virtual human video generation framework.
Our main contributions could be summarized as follows:
- The released model can generate dance videos of the human character in a reference image under the given pose sequence. The result quality exceeds almost all current open source models within the same topic.
- We release the
pose alignalgorithm so that users could align arbitrary dance videos to arbitrary reference images, which SIGNIFICANTLY improved inference performance and enhanced model usability. - We have fixed several important bugs and made some improvement based on the code of Moore-AnimateAnyone.
demo.0.mp4 |
demo.1.mp4 |
demo.2.mp4 |
demo.3.mp4 |
demo.4.mp4 |
demo.5.mp4 |
demo.6.mp4 |
demo.7.mp4 |
- [05/27/2024] Release
MusePoseand pretrained models. - [05/31/2024] Support Comfyui-MusePose
- [06/14/2024] Bug Fixed in
inference_v2.yaml.
- release our trained models and inference codes of MusePose.
- release pose align algorithm.
- Comfyui-MusePose
- training guidelines.
- Huggingface Gradio demo.
- a improved architecture and model (may take longer).
We provide a detailed tutorial about the installation and the basic usage of MusePose for new users:
To prepare the Python environment and install additional packages such as opencv, diffusers, mmcv, etc., please follow the steps below:
We recommend a python version >=3.10 and cuda version =11.7. Then build environment as follows:
pip install -r requirements.txtpip install --no-cache-dir -U openmim
mim install mmengine
mim install "mmcv>=2.0.1"
mim install "mmdet>=3.1.0"
mim install "mmpose>=1.1.0" You can download weights manually as follows:
-
Download our trained weights.
-
Download the weights of other components:
- sd-image-variations-diffusers
- sd-vae-ft-mse
- dwpose
- yolox - Make sure to rename to
yolox_l_8x8_300e_coco.pth - image_encoder
Finally, these weights should be organized in pretrained_weights as follows:
./pretrained_weights/
|-- MusePose
| |-- denoising_unet.pth
| |-- motion_module.pth
| |-- pose_guider.pth
| └── reference_unet.pth
|-- dwpose
| |-- dw-ll_ucoco_384.pth
| └── yolox_l_8x8_300e_coco.pth
|-- sd-image-variations-diffusers
| └── unet
| |-- config.json
| └── diffusion_pytorch_model.bin
|-- image_encoder
| |-- config.json
| └── pytorch_model.bin
└── sd-vae-ft-mse
|-- config.json
└── diffusion_pytorch_model.bin
Prepare your referemce images and dance videos in the folder ./assets and organnized as the example:
./assets/
|-- images
| └── ref.png
└── videos
└── dance.mp4
Get the aligned dwpose of the reference image:
python pose_align.py --imgfn_refer ./assets/images/ref.png --vidfn ./assets/videos/dance.mp4
After this, you can see the pose align results in ./assets/poses, where ./assets/poses/align/img_ref_video_dance.mp4 is the aligned dwpose and the ./assets/poses/align_demo/img_ref_video_dance.mp4 is for debug.
Add the path of the reference image and the aligned dwpose to the test config file ./configs/test_stage_2.yaml as the example:
test_cases:
"./assets/images/ref.png":
- "./assets/poses/align/img_ref_video_dance.mp4"
Then, simply run
python test_stage_2.py --config ./configs/test_stage_2.yaml
./configs/test_stage_2.yaml is the path to the inference configuration file.
Finally, you can see the output results in ./output/
If you want to reduce the VRAM cost, you could set the width and height for inference. For example,
python test_stage_2.py --config ./configs/test_stage_2.yaml -W 512 -H 512
It will generate the video at 512 x 512 first, and then resize it back to the original size of the pose video.
Currently, it takes 16GB VRAM to run on 512 x 512 x 48 and takes 28GB VRAM to run on 768 x 768 x 48. However, it should be noticed that the inference resolution would affect the final results (especially face region).
If you want to enhance the face region to have a better consistency of the face, you could use FaceFusion. You could use the face-swap function to swap the face in the reference image to the generated video.
- We thank AnimateAnyone for their technical report, and have refer much to Moore-AnimateAnyone and diffusers.
- We thank open-source components like AnimateDiff, dwpose, Stable Diffusion, etc..
Thanks for open-sourcing!
- Detail consitency: some details of the original character are not well preserved (e.g. face region and complex clothing).
- Noise and flickering: we observe noise and flicking in complex background.
@article{musepose,
title={MusePose: a Pose-Driven Image-to-Video Framework for Virtual Human Generation},
author={Tong, Zhengyan and Li, Chao and Chen, Zhaokang and Wu, Bin and Zhou, Wenjiang},
journal={arxiv},
year={2024}
}code: The code of MusePose is released under the MIT License. There is no limitation for both academic and commercial usage.model: The trained model are available for non-commercial research purposes only.other opensource model: Other open-source models used must comply with their license, such asft-mse-vae,dwpose, etc..- The testdata are collected from internet, which are available for non-commercial research purposes only.
AIGC: This project strives to impact the domain of AI-driven video generation positively. Users are granted the freedom to create videos using this tool, but they are expected to comply with local laws and utilize it responsibly. The developers do not assume any responsibility for potential misuse by users.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent