The CC-OCR dataset was jointly developed by Alibaba Group, Huazhong University of Science and Technology, and South China University of Technology in 2024 to provide a comprehensive and challenging benchmark for evaluating the performance of large multimodal models in text recognition (OCR) tasks.CC-OCR: A Comprehensive and Challenging OCR Benchmark for Evaluating Large Multimodal Models in Literacy". The dataset covers four core tasks: multi-scene text reading, multi-language text reading, document parsing, and key information extraction, and contains 39 subsets and 7,058 fully annotated images. The launch of CC-OCR fills the gap in the evaluation of current multimodal models in terms of complex structures and fine-grained visual challenges, and is of great significance to promoting the progress of multimodal models in practical applications.

Citation

@misc{yang2024ccocr,
title={CC-OCR: A Comprehensive and Challenging OCR Benchmark for Evaluating Large Multimodal Models in Literacy},
author={Zhibo Yang and Jun Tang and Zhaohai Li and Pengfei Wang and Jianqiang Wan and Humen Zhong and Xuejing Liu and Mingkun Yang and Peng Wang and Shuai Bai and LianWen Jin and Junyang Lin},
year={2024},
eprint={2412.02210},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2412.02210},
}