Kebii / R2ET

(CVPR 2023) Skinned Motion Retargeting with Residual Perception of Motion Semantics & Geometry
MIT License
174 stars 12 forks source link

Skinned Motion Retargeting with Residual Perception of Motion Semantics & Geometry

This is the code for the CVPR 2023 paper Skinned Motion Retargeting with Residual Perception of Motion Semantics & Geometry by Jiaxu Zhang, et al.

R2ET is a neural motion retargeting model that can preserve the source motion semantics and avoid interpenetration in the target motion.

Quick Start

1. Conda environment

conda create python=3.9 --name r2et
conda activate r2et

2. Install dependencies (Anaconda installation is recommended)

3. Download and Install Blender

4. Data preparation

Training data:

5. Install the CUDA implementation of SDF function

Inference

Performing inference using bvh files:

python3 inference_bvh.py --config ./config/inference_bvh_cfg.yaml

Training

Skeleton-aware Network:

python3 train_skeleton_aware.py --config ./config/train_skeleton_aware.yaml

Shape-aware Network:

python3 train_shape_aware.py --config ./config/train_shape_aware.yaml

Visualization

The visualization parameters are in the ./visualization/options.py

cd ./visualization
blender -P visualize.py

Citation

Acknowledgments

Thanks to PMnet, SAN and NKN, our code is partially borrowing from them.