- Task 1 - Text Localization
- Task 2 - Script identification
- Task 3 - Joint text detection and script identification
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.
| Date | Method | Hmean | Precision | Recall | Average Precision | |||
|---|---|---|---|---|---|---|---|---|
| 2020-04-16 | TH | 55.77% | 41.01% | 87.12% | 73.23% | |||
| 2019-11-08 | Sogou_OCR | 53.96% | 39.23% | 86.38% | 68.24% | |||
| 2020-04-22 | AntAI-Cognition | 52.67% | 37.81% | 86.78% | 68.83% | |||
| 2019-08-08 | JDAI | 49.37% | 34.52% | 86.65% | 74.80% | |||
| 2019-05-30 | PMTD | 48.30% | 33.75% | 84.95% | 72.61% | |||
| 2019-06-02 | NJU-ImagineLab | 47.22% | 32.46% | 86.61% | 68.80% | |||
| 2019-03-29 | GNNets (single scale) | 47.15% | 33.90% | 77.45% | 51.73% | |||
| 2019-05-08 | Baidu-VIS | 46.60% | 32.89% | 79.96% | 26.26% | |||
| 2019-03-23 | PMTD | 45.64% | 32.09% | 78.99% | 52.51% | |||
| 2018-11-20 | Pixel-Anchor | 44.36% | 32.15% | 71.54% | 27.03% | |||
| 2019-06-11 | 4Paradigm-Data-Intelligence | 43.11% | 29.07% | 83.37% | 24.42% | |||
| 2019-05-23 | 4Paradigm-Data-Intelligence | 42.96% | 29.06% | 82.32% | 24.00% | |||
| 2017-06-28 | SCUT_DLVClab1 | 42.20% | 29.86% | 71.92% | 52.71% | |||
| 2018-10-29 | Amap-CVLab | 41.86% | 28.52% | 78.63% | 51.64% | |||
| 2019-12-13 | BDN | 41.08% | 26.95% | 86.32% | 23.12% | |||
| 2018-11-28 | CRAFT | 40.43% | 28.37% | 70.34% | 19.96% | |||
| 2017-11-09 | EAST++ | 40.08% | 27.29% | 75.47% | 34.10% | |||
| 2018-05-18 | PSENet_NJU_ImagineLab (single-scale) | 39.63% | 27.08% | 73.87% | 20.21% | |||
| 2018-11-15 | USTC-NELSLIP | 38.09% | 24.99% | 80.04% | 46.23% | |||
| 2019-01-08 | ALGCD_CP | 36.41% | 23.88% | 76.63% | 32.94% | |||
| 2018-12-04 | SPCNet_TongJi & UESTC (multi scale) | 35.96% | 23.95% | 72.19% | 17.42% | |||
| 2018-12-05 | EPTN-SJTU | 33.30% | 21.52% | 73.52% | 33.46% | |||
| 2019-05-30 | Thesis-SE | 32.77% | 21.27% | 71.39% | 30.12% | |||
| 2018-03-12 | ATL Cangjie OCR | 32.11% | 20.84% | 69.98% | 35.50% | |||
| 2019-07-15 | stela | 32.07% | 22.31% | 57.05% | 28.65% | |||
| 2017-06-29 | SARI_FDU_RRPN_v1 | 30.15% | 19.23% | 69.71% | 40.70% | |||
| 2018-12-03 | SPCNet_TongJi & UESTC (single scale) | 27.97% | 17.16% | 75.68% | 13.15% | |||
| 2017-06-28 | SARI_FDU_RRPN_v0 | 25.27% | 15.60% | 66.48% | 36.44% | |||
| 2017-06-30 | TH-DL | 25.09% | 16.98% | 48.08% | 20.81% | |||
| 2017-06-30 | linkage-ER-Flow | 17.34% | 10.30% | 54.72% | 12.37% | |||
| 2017-06-30 | Sensetime OCR | 14.98% | 8.27% | 80.04% | 40.38% |