Skip to content

dongludeeplearning/SignAvatar

Repository files navigation

SignAvatar Project

SignAvatar is a transformer-based framework that reconstructs and generates expressive 3D sign language motions at the word level, enhanced by curriculum learning and supported by our newly introduced ASL3DWord dataset.

🔗 Paper: SignAvatar: Sign Language 3D Motion Reconstruction and Generation
This repository contains the official implementation of the above paper.


📁 Project Structure

SignAvatar/
├── blender_app/                    # Blender visualization code
├── checkpoints/                    # Model checkpoints - Download separately
├── dataset/                 
│   ├── data/                      # Dataset files - Download separately
│   │   ├── ASL3DWord/         
│   │   │   ├── test/
│   │   │   └── train/
│   │   ├── word_projection/       # Word projection mappings
│   │   └── WLASL_v0.3.json       
│   ├── dataset.py                 # Dataset classes and utilities
│   ├── ......
├── evaluate/                       # Model evaluation and metrics
├── generate/                       # Sequence generation scripts
├── models/                         # Model architectures
│   ├── architectures/            
│   ├── modeltype/                 
│   ├── smplx/                     # SMPLX body - Download separately
│   │   ├── SMPLX_NEUTRAL.npz     
│   │   └── kin_pose53_smplx.pkl   
│   ├── ......
......

📥 Request Dataset

1. ASL3DWord Dataset

mkdir -p dataset/data/ASL3DWord

To request access to the ASL3DWord dataset, please send your request via email: [email protected]

When sending your request, kindly include the following information:
- Name
- Institution / Organization
- Research Purpose
- Which resource are you requesting (Dataset / Checkpoints / Both)
- Statement of Agreement: I confirm that this resource will be used for research and educational purposes only.

2. SMPLX Models

Required files:
 - SMPLX_NEUTRAL.npz files 
 - kin_pose53_smplx.pkl 

# Download SMPLX_NEUTRAL from: https://smpl-x.is.tue.mpg.de/
# Download kin_pose53_smplx.pkl from previous link

🚀 Quick Start

Installation

# 1. Clone the repository
git clone https://github.com/dongludeeplearning/SignAvatar.git
cd SignAvatar

# 2. Create conda environment
conda env create -f environment.yaml
conda activate signavatar

# 3. Download required files (see section above)
# 4. Verify installation
python -c "import torch; print('PyTorch version:', torch.__version__)"

Training

# Train CVAE model
bash run_train_cvae.sh

# Train STGCN model  
bash run_train_stgcn.sh

Generation

# Generate pose sequences
bash run_generation.sh

# Generate 3D meshes
python generate/generate_sequences_mesh.py

Evaluation

# Evaluate CVAE model
bash run_evaluate_cvae.sh

📖 Citation

If you use this code or dataset in your research, please cite our accompanying paper:

@inproceedings{dong2024signavatar,
  title={Signavatar: Sign language 3d motion reconstruction and generation},
  author={Dong, Lu and Chaudhary, Lipisha and Xu, Fei and Wang, Xiao and Lary, Mason and Nwogu, Ifeoma},
  booktitle={2024 IEEE 18th International Conference on Automatic Face and Gesture Recognition (FG)},
  pages={1--10},
  year={2024},
  organization={IEEE}
}

📜 License

This project is released under the CC BY-NC 4.0 License — for research and educational use only.

About

Official code release for the paper SignAvatar: Sign Language 3D Motion Reconstruction and Generation in 2024 IEEE FG Conference.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors