metadata

license: mit
task_categories:
  - image-to-3d
tags:
  - hand model
  - 3d reconstruction
  - mano
  - synthetic
  - textures
  - accessories
size_categories:
  - 100K<n<1M
language:
  - en
pretty_name: DART

DART: Articulated Hand Model with Diverse Accessories and Rich Textures

Daiheng Gao^* · Yuliang Xiu^* · Kailin Li^* · Lixin Yang^*
Feng Wang · Peng Zhang · Bang Zhang · Cewu Lu · Ping Tan

NeurIPS 2022 (Datasets and Benchmarks Track)

Update

[2022.10.07] DART's raw textures+accessories are released at RAW
[2022.09.29] DART Unity GUI's source code is publicly available at GUI.

Environment

numpy
cv2
imageio
PyTorch
PyTorch3D (>= 0.6)
manotorch

Data

Please download the data from HuggingFace/Dataset or Baidu Pan (4w3r) and put them in the data/DARTset folder.

git clone https://huggingface.co/datasets/yuliang/dart data/DARTset

Then download MANO from the official website and put it in the assets folder.

Your directory should look like this:

.
├── DARTset.py
├── DARTset_utils.py
├── assets
│   └── mano_v1_2
├── data
│   └── DARTset
│       ├── train
│       │   ├── 0
│       │   ├── 0_wbg
│       │   ├── part_0.pkl
│       │   |-- ...
│       └── test

Visualization

python DARTset.py

You can modify this line in DARTset.py to change the train/test data split.

Post Processing with Unity GUI

Please check postprocess folder to learn how to generate intermediate output using DART's Unity GUI.

Citation

If you find our work useful in your research, please cite:

@inproceedings{gao2022dart,
    title={{DART: Articulated Hand Model with Diverse Accessories and Rich Textures}},
    author={Daiheng Gao and Yuliang Xiu and Kailin Li and Lixin Yang and Feng Wang and Peng Zhang and Bang Zhang and Cewu Lu and Ping Tan},
    booktitle={Thirty-sixth Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
    year={2022},
}