method: TH2020-04-16

Authors: Tsinghua University and Hyundai Motor Group AIRS Company

Email: Shanyu Xiao: xiaosy19@mails.tsinghua.edu.cn

Description: We have built an end-to-end scene text spotter based on Mask R-CNN & Transformer. The ResNeXt-101 backbone and multiscale training/testing are used.

method: Sogou_OCR2019-11-08

Authors: Xudong Rao, Lulu Xu, Long Ma, Xuefeng Su

Description: An arbitrary-shaped text detection method based on Mask R-CNN, we use resnext-152 as our backbone, multi-scale training and testing are adopted to get the final results.

method: AntAI-Cognition2020-04-22

Authors: Qingpei Guo, Yudong Liu, Pengcheng Yang, Yonggang Li, Yongtao Wang, Jingdong Chen, Wei Chu

Affiliation: Ant Group & PKU

Email: qingpei.gqp@antgroup.com

Description: We are from Ant Group & PKU. Our approach is an ensemble method with three text detection models. The text detection models mainly follow the MaskRCNN framework[1], with different backbones(ResNext101-64x4d[2], CBNet[3], ResNext101-32x32d_wsl[4]) used. GBDT[5] is trained to normalize confidence scores and select quadrilateral boxes with the highest quality from all text detection models' outputs. Multi-scale training and testing are adopted for all basic models. For the training set, we also add ICDAR19 MLT datasets, both training & validation sets are used to get the final result.

[1] He K, Gkioxari G, Dollár P, et al. Mask r-cnn[C]//Proceedings of the IEEE international conference on computer vision. 2017: 2961-2969. [2] Xie S, Girshick R, Dollár P, et al. Aggregated residual transformations for deep neural networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 1492-1500. [3] Liu Y, Wang Y, Wang S, et al. Cbnet: A novel composite backbone network architecture for object detection[J]. arXiv preprint arXiv:1909.03625, 2019. [4] Mahajan D, Girshick R, Ramanathan V, et al. Exploring the limits of weakly supervised pretraining[C]//Proceedings of the European Conference on Computer Vision (ECCV). 2018: 181-196. [5] Ke G, Meng Q, Finley T, et al. Lightgbm: A highly efficient gradient boosting decision tree[C]//Advances in neural information processing systems. 2017: 3146-3154.

Ranking Table

Description Paper Source Code
DateMethodHmeanPrecisionRecallAverage Precision
2020-04-16TH59.17%44.09%89.91%79.49%
2019-11-08Sogou_OCR57.05%42.33%87.46%68.55%
2020-04-22 AntAI-Cognition56.68%41.26%90.51%74.90%
2019-03-29GNNets (single scale)52.80%39.41%79.97%47.89%
2018-11-20Pixel-Anchor52.70%39.01%81.16%42.19%
2019-08-08JDAI52.31%37.11%88.60%73.66%
2019-05-30PMTD51.69%36.90%86.29%75.91%
2019-05-08Baidu-VIS51.66%37.18%84.62%30.96%
2019-03-23PMTD51.22%36.71%84.73%69.90%
2019-06-02NJU-ImagineLab50.41%35.04%89.81%75.59%
2019-06-11 4Paradigm-Data-Intelligence47.35%32.46%87.48%28.06%
2019-05-234Paradigm-Data-Intelligence47.34%32.51%87.05%28.48%
2018-10-29Amap-CVLab46.39%32.63%80.22%66.26%
2019-07-15stela46.09%32.67%78.26%57.95%
2018-05-18PSENet_NJU_ImagineLab (single-scale)45.98%32.86%76.53%25.15%
2017-11-09EAST++44.21%30.80%78.34%27.79%
2019-12-13BDN43.62%29.16%86.52%25.00%
2018-11-28CRAFT43.46%31.91%68.09%22.40%
2018-11-15USTC-NELSLIP43.44%29.18%84.97%69.13%
2017-06-28SCUT_DLVClab143.07%31.35%68.79%47.05%
2018-12-04 SPCNet_TongJi & UESTC (multi scale)41.71%28.26%79.62%22.37%
2018-03-12ATL Cangjie OCR40.05%26.58%81.19%64.38%
2019-01-08ALGCD_CP39.92%26.57%80.24%24.09%
2018-12-05EPTN-SJTU38.53%25.66%77.32%21.78%
2019-05-30Thesis-SE37.74%25.09%76.08%21.23%
2017-06-29SARI_FDU_RRPN_v134.74%22.99%71.06%51.15%
2018-12-03SPCNet_TongJi & UESTC (single scale)31.65%19.72%80.12%15.98%
2017-06-28SARI_FDU_RRPN_v029.58%18.86%68.59%35.58%
2017-06-30TH-DL29.11%20.54%49.94%30.05%
2017-06-30Sensetime OCR16.23%9.03%80.09%47.00%
2017-06-30linkage-ER-Flow4.81%2.95%13.05%0.99%

Ranking Graphic

Ranking Graphic