method: SituTech_OCR2021-03-11

Authors: Kui Lyu, Chuanhe Liu

Affiliation: Beijing Situ Vision Technologies Co. Ltd

Email: lvkui@situdata.com

Description: In this work, we design an elegant text detection model. Our detector is similar to DBNet, but there are some difference. More specifically, we have introduced an advanced detector backbone, a classic network EfficientDet, with flexible scales and stronger ability to extract features. Another breakthrough is that we optimized the label generation strategy in DBNet. In the original work, the positive area generation and the expansion of the positive area to the bounding box used the Vatti clipping algorithm, which is less robust with different area perimeter ratios. We optimized this function to make the label transform between positive area and bounding box more reasonable.

If you have any questions, please contact us.
SituAIgorithm Team, Beijing Situ Vision Technologies Co. Ltd

Authors: USTB&Tele-Yizhi

Description: Our detector is based on TextBPN++, and pretrained on about two million images. We finetune on preprocessed MLT, and then Multi-scale testing is used to get the result.

method: CPN (multi-scale)2024-05-30

Authors: Longhuang Wu, Shangxuan Tian, Youxin Wang, Pengfei Xiong

Email: wlonghuang@gmail.com

Description: We propose a Complementary Proposal Network (CPN) that seamlessly and parallelly integrates semantic and geometric information for superior performance. This Result is achieved with single Swin-L backbone and multi-scale testing policy. No model ensemble is used.

Ranking Table

Description Paper Source Code
DateMethodHmeanPrecisionRecallAverage Precision
2021-03-11SituTech_OCR62.45%54.60%72.93%39.06%
2025-06-30USTB&Tele-Yizhi TextBPN-MLOCR (multi-scale)59.26%47.65%78.33%36.09%
2024-05-30CPN (multi-scale)58.98%47.28%78.37%58.25%
2019-06-04multi-stage_text_detector_v456.37%45.16%74.99%35.34%
2019-05-30PMTD53.02%42.11%71.56%49.26%
2019-05-31A two-stage text detector based on cascade rcnn46.31%36.13%64.47%40.73%
2019-06-02A two-stage text detector based on cascade rcnn(using total 10000 images of mlt19)45.75%35.15%65.54%40.48%
2019-05-26two stage text detector42.58%33.37%58.83%34.28%
2019-06-03CRAFTS42.10%36.28%50.15%21.36%
2019-05-28CRAFTS(Initial)38.98%31.03%52.41%17.55%
2019-05-24PSENet_v134.47%27.67%45.69%22.64%

Ranking Graphic

Ranking Graphic