Skip to content

Latest commit

 

History

History
18 lines (14 loc) · 687 Bytes

README.md

File metadata and controls

18 lines (14 loc) · 687 Bytes

Chinese Chess Detection Dataset

  • Use AnyLabeling to annotate all data in data/data_01 and data/data_02.
  • Run make_data.sh to generate the training data.
conda create -n chessai-dataprep python=3.9
conda activate chessai-dataprep
pip install -r requirements.txt
bash make_data.sh

Citation

Please cite this paper if it helps your research:

@software{Nguyen_ChessAI_Data_-_2023,author = {Nguyen, Viet-Anh},doi = {10.5281/zenodo.1234},month = nov,title = {{ChessAI Data - Data for Chinese chess game position recognition}},url = {https://github.com/nrl-ai/chessai-data},version = {1.0.0},year = {2023}}