Reproduce “An open dataset for the evolution of oracle bone characters: EVOBC” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Haisu Guan, Jinpeng Wan, Yuliang Liu, Pengjie Wang, Kaile Zhang, Zhebin Kuang, Xinyu Wang, Xiang Bai and 1 more
The authors introduce EVOBC, a labeled image dataset for comparing oracle-bone characters with later forms of Chinese writing. It contains 229,170 images across 13,714 categories, with each character represented in up to six historical periods. On a 9:1 training-validation classification split, Swin Transformer v2 achieves 86.66% Top-1 and 97.93% Top-20 accuracy; preliminary deciphering results remain limited.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.