- 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 | 58.66% | 49.57% | 71.82% | 49.46% | |||
2019-11-08 | Sogou_OCR | 56.69% | 47.66% | 69.94% | 47.27% | |||
2020-04-22 | AntAI-Cognition | 56.55% | 46.52% | 72.10% | 46.56% | |||
2019-05-08 | Baidu-VIS | 53.38% | 42.87% | 70.72% | 29.94% | |||
2019-05-30 | PMTD | 53.34% | 42.54% | 71.51% | 49.93% | |||
2019-08-08 | JDAI | 53.26% | 42.56% | 71.13% | 50.82% | |||
2019-06-02 | NJU-ImagineLab | 52.80% | 41.06% | 73.94% | 49.97% | |||
2019-03-23 | PMTD | 50.87% | 40.87% | 67.37% | 45.30% | |||
2019-06-11 | 4Paradigm-Data-Intelligence | 49.41% | 37.84% | 71.18% | 26.31% | |||
2019-05-23 | 4Paradigm-Data-Intelligence | 48.88% | 37.61% | 69.78% | 25.79% | |||
2018-11-20 | Pixel-Anchor | 47.93% | 40.71% | 58.24% | 22.48% | |||
2019-03-29 | GNNets (single scale) | 46.72% | 38.47% | 59.46% | 30.88% | |||
2018-11-28 | CRAFT | 46.15% | 37.37% | 60.33% | 22.35% | |||
2019-12-13 | BDN | 46.05% | 34.06% | 71.03% | 23.70% | |||
2018-10-29 | Amap-CVLab | 44.87% | 35.48% | 61.00% | 30.08% | |||
2018-11-15 | USTC-NELSLIP | 44.42% | 32.85% | 68.55% | 38.69% | |||
2017-11-09 | EAST++ | 43.15% | 33.57% | 60.37% | 27.28% | |||
2018-05-18 | PSENet_NJU_ImagineLab (single-scale) | 41.03% | 31.96% | 57.29% | 17.80% | |||
2018-12-04 | SPCNet_TongJi & UESTC (multi scale) | 40.84% | 31.29% | 58.81% | 17.97% | |||
2019-01-08 | ALGCD_CP | 40.45% | 30.10% | 61.65% | 26.49% | |||
2019-07-15 | stela | 39.20% | 31.46% | 51.99% | 25.52% | |||
2018-03-12 | ATL Cangjie OCR | 38.91% | 28.76% | 60.12% | 31.21% | |||
2017-06-28 | SCUT_DLVClab1 | 37.02% | 31.48% | 44.93% | 25.34% | |||
2019-05-30 | Thesis-SE | 34.72% | 25.80% | 53.07% | 21.64% | |||
2018-12-05 | EPTN-SJTU | 34.48% | 25.57% | 52.91% | 21.71% | |||
2018-12-03 | SPCNet_TongJi & UESTC (single scale) | 30.87% | 21.16% | 57.04% | 11.89% | |||
2017-06-29 | SARI_FDU_RRPN_v1 | 30.72% | 22.58% | 48.02% | 19.88% | |||
2017-06-28 | SARI_FDU_RRPN_v0 | 28.73% | 19.91% | 51.53% | 24.29% | |||
2017-06-30 | TH-DL | 20.20% | 16.53% | 25.97% | 9.24% | |||
2017-06-30 | Sensetime OCR | 18.68% | 10.93% | 64.03% | 27.49% | |||
2017-06-30 | linkage-ER-Flow | 18.52% | 12.13% | 39.18% | 6.15% |