Hallo: Hierarchical Audio-Driven Visual Synthesis for Portrait Image Animation

Mingwang Xu^1*  Hui Li^1*  Qingkun Su^1*  Hanlin Shang¹  Liwei Zhang¹  Ce Liu³ 

Jingdong Wang²  Yao Yao⁴  Siyu Zhu¹ 

¹Fudan University  ²Baidu Inc  ³ETH Zurich  ⁴Nanjing University

📸 Showcase

head.mp4

🎬 Honoring Classic Films

Devil Wears Prada	Green Book	Infernal Affairs
Patch Adams	Tough Love	Shawshank Redemption

Explore more examples.

📰 News

• 2024/06/15: ✨✨✨ Released some images and audios for inference testing on 🤗Huggingface.
• 2024/06/15: 🎉🎉🎉 Launched the first version on 🫡GitHub.

🤝 Community Resources

Explore the resources developed by our community to enhance your experience with Hallo:

• Demo on Huggingface - Check out this easy-to-use Gradio demo by @multimodalart.
• hallo-webui - Explore the WebUI created by @daswer123.
• hallo-for-windows - Utilize Hallo on Windows with the guide by @sdbds.
• ComfyUI-Hallo - Integrate Hallo with the ComfyUI tool by @AIFSH.

Thanks to all of them.

Join our community and explore these amazing resources to make the most out of Hallo. Enjoy and elevate their creative projects!

🔧️ Framework

⚙️ Installation

• System requirement: Ubuntu 20.04/Ubuntu 22.04, Cuda 12.1
• Tested GPUs: A100

Create conda environment:

  conda create -n hallo python=3.10
  conda activate hallo

Install packages with pip

  pip install -r requirements.txt
  pip install .

Besides, ffmpeg is also needed:

  apt-get install ffmpeg

🗝️️ Usage

The entry point for inference is scripts/inference.py. Before testing your cases, two preparations need to be completed:

1. Download all required pretrained models.
2. Prepare source image and driving audio pairs.
3. Run inference.

📥 Download Pretrained Models

You can easily get all pretrained models required by inference from our HuggingFace repo.

Clone the pretrained models into ${PROJECT_ROOT}/pretrained_models directory by cmd below:

git lfs install
git clone https://huggingface.co/fudan-generative-ai/hallo pretrained_models

Or you can download them separately from their source repo:

• hallo: Our checkpoints consist of denoising UNet, face locator, image & audio proj.
• audio_separator: Kim_Vocal_2 MDX-Net vocal removal model. (Thanks to KimberleyJensen)
• insightface: 2D and 3D Face Analysis placed into pretrained_models/face_analysis/models/. (Thanks to deepinsight)
• face landmarker: Face detection & mesh model from mediapipe placed into pretrained_models/face_analysis/models.
• motion module: motion module from AnimateDiff. (Thanks to guoyww).
• sd-vae-ft-mse: Weights are intended to be used with the diffusers library. (Thanks to stablilityai)
• StableDiffusion V1.5: Initialized and fine-tuned from Stable-Diffusion-v1-2. (Thanks to runwayml)
• wav2vec: wav audio to vector model from Facebook.

Finally, these pretrained models should be organized as follows:

./pretrained_models/
|-- audio_separator/
|   `-- Kim_Vocal_2.onnx
|-- face_analysis/
|   `-- models/
|       |-- face_landmarker_v2_with_blendshapes.task  # face landmarker model from mediapipe
|       |-- 1k3d68.onnx
|       |-- 2d106det.onnx
|       |-- genderage.onnx
|       |-- glintr100.onnx
|       `-- scrfd_10g_bnkps.onnx
|-- motion_module/
|   `-- mm_sd_v15_v2.ckpt
|-- sd-vae-ft-mse/
|   |-- config.json
|   `-- diffusion_pytorch_model.safetensors
|-- stable-diffusion-v1-5/
|   |-- feature_extractor/
|   |   `-- preprocessor_config.json
|   |-- model_index.json
|   |-- unet/
|   |   |-- config.json
|   |   `-- diffusion_pytorch_model.safetensors
|   `-- v1-inference.yaml
`-- wav2vec/
    |-- wav2vec2-base-960h/
    |   |-- config.json
    |   |-- feature_extractor_config.json
    |   |-- model.safetensors
    |   |-- preprocessor_config.json
    |   |-- special_tokens_map.json
    |   |-- tokenizer_config.json
    |   `-- vocab.json

🛠️ Prepare Inference Data

Hallo has a few simple requirements for input data:

For the source image:

1. It should be cropped into squares.
2. The face should be the main focus, making up 50%-70% of the image.
3. The face should be facing forward, with a rotation angle of less than 30° (no side profiles).

For the driving audio:

1. It must be in WAV format.
2. It must be in English since our training datasets are only in this language.
3. Ensure the vocals are clear; background music is acceptable.

We have provided some samples for your reference.

🎮 Run Inference

Simply to run the scripts/inference.py and pass source_image and driving_audio as input:

python scripts/inference.py --source_image examples/reference_images/1.jpg --driving_audio examples/driving_audios/1.wav

Animation results will be saved as ${PROJECT_ROOT}/.cache/output.mp4 by default. You can pass --output to specify the output file name. You can find more examples for inference at examples folder.

For more options:

usage: inference.py [-h] [-c CONFIG] [--source_image SOURCE_IMAGE] [--driving_audio DRIVING_AUDIO] [--output OUTPUT] [--pose_weight POSE_WEIGHT]
                    [--face_weight FACE_WEIGHT] [--lip_weight LIP_WEIGHT] [--face_expand_ratio FACE_EXPAND_RATIO]

options:
  -h, --help            show this help message and exit
  -c CONFIG, --config CONFIG
  --source_image SOURCE_IMAGE
                        source image
  --driving_audio DRIVING_AUDIO
                        driving audio
  --output OUTPUT       output video file name
  --pose_weight POSE_WEIGHT
                        weight of pose
  --face_weight FACE_WEIGHT
                        weight of face
  --lip_weight LIP_WEIGHT
                        weight of lip
  --face_expand_ratio FACE_EXPAND_RATIO
                        face region

📅️ Roadmap

Status	Milestone	ETA
✅	Inference source code meet everyone on GitHub	2024-06-15
✅	Pretrained models on Huggingface	2024-06-15
🚧	Optimizing Inference Performance	2024-06-23
🚧	Optimizing Performance on images with a resolution of 256x256.	2024-06-23
🚀	Improving the model's performance on Mandarin Chinese	2024-06-25
🚀	Releasing data preparation and training scripts	2024-06-28

Other Enhancements

• Enhancement: Test and ensure compatibility with Windows operating system. #39
• Bug: Output video may lose several frames. #41
• Bug: Sound volume affecting inference results (audio normalization).
• Enhancement: Inference code logic optimization.
• Enhancement: Enhancing performance on low resolutions(256x256) to support more efficient usage.

📝 Citation

If you find our work useful for your research, please consider citing the paper:

@misc{xu2024hallo,
  title={Hallo: Hierarchical Audio-Driven Visual Synthesis for Portrait Image Animation},
  author={Mingwang Xu and Hui Li and Qingkun Su and Hanlin Shang and Liwei Zhang and Ce Liu and Jingdong Wang and Yao Yao and Siyu zhu},
  year={2024},
  eprint={2406.08801},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}

🌟 Opportunities Available

Multiple research positions are open at the Generative Vision Lab, Fudan University! Include:

• Research assistant
• Postdoctoral researcher
• PhD candidate
• Master students

Interested individuals are encouraged to contact us at [email protected] for further information.

⚠️ Social Risks and Mitigations

The development of portrait image animation technologies driven by audio inputs poses social risks, such as the ethical implications of creating realistic portraits that could be misused for deepfakes. To mitigate these risks, it is crucial to establish ethical guidelines and responsible use practices. Privacy and consent concerns also arise from using individuals' images and voices. Addressing these involves transparent data usage policies, informed consent, and safeguarding privacy rights. By addressing these risks and implementing mitigations, the research aims to ensure the responsible and ethical development of this technology.

👏 Community Contributors

Thank you to all the contributors who have helped to make this project better!