Skip to content

Latest commit

 

History

History
executable file
·
43 lines (31 loc) · 1.06 KB

README.md

File metadata and controls

executable file
·
43 lines (31 loc) · 1.06 KB

Shrec24

[SHREC24] Skeleton-based Self-Supervised Learning For Dynamic Hand Gesture Recognition

Check out the SHREC24 paper here.

hippo

Updates

  • ...

Installation

Create and activate conda environment:

conda create -n shrec24 python=3.10
conda activate shrec24

Install all dependencies:

pip install -r requirements.txt

Dataset

Download the SHREC'24 dataset. And set the path to training and test data in the shrec24_config.yaml file in the configs/ folder.

 train_data_dir: './path/to/Train-set/'
 test_data_dir: './path/to/Test-set/'

Training

bash train.sh --config_file configs/shrec24_config.yaml

Evaluation

bash eval.sh --config_file configs/shrec24_config.yaml

We thank MAE and STGCN for making their code available