UniCalli Dataset
A large-scale Chinese calligraphy dataset with dense annotations, built for the UniCalli project.
Overview
The dataset contains Chinese calligraphy images collected from historical works spanning multiple dynasties, covering 95+ calligraphy masters and 5 major script styles. Each sample is annotated with bounding boxes, modern text transcriptions, author attribution, and font style labels.
Webpage
- Project Page: https://envision-research.github.io/UniCalli/
- GitHub: https://github.com/EnVision-Research/UniCalli
- HuggingFace Model: https://huggingface.co/TSXu/Unicalli_Pro
- HuggingFace Demo: https://huggingface.co/spaces/TSXu/UniCalli_Dev
- ModelScope: https://www.modelscope.cn/models/tianshuo/UniCalli-base
- arXiv: https://arxiv.org/abs/2510.13745
Citation
If you use this dataset in your research, please cite:
@article{xu2025unicalli,
title={UniCalli: A Unified Diffusion Framework for Column-Level Generation and Recognition of Chinese Calligraphy},
author={Xu, Tianshuo and Wang, Kai and Chen, Zhifei and Wu, Leyi and Wen, Tianshui and Chao, Fei and Chen, Ying-Cong},
journal={arXiv preprint arXiv:2025.13745},
year={2025}
}