- Task 1 - Text Localization
- Task 2 - Script identification
- Task 3 - Joint text detection and script identification
method: FPDIoU2024-04-02
Authors: Siliang Ma
Affiliation: South China University of Technology
Description: hhh
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.
Description Paper Source Code
| Date | Method | Hmean | Precision | Recall | Average Precision | |||
|---|---|---|---|---|---|---|---|---|
| 2024-04-02 | FPDIoU | 45.35% | 32.16% | 76.86% | 24.25% | |||
| 2020-04-16 | TH | 44.92% | 29.49% | 94.22% | 75.64% | |||
| 2019-11-08 | Sogou_OCR | 44.89% | 29.13% | 97.76% | 85.12% | |||
| 2023-05-22 | DeepSolo++ (ResNet-50) | 43.09% | 28.14% | 91.91% | 84.32% | |||
| 2020-04-22 | AntAI-Cognition | 42.78% | 27.46% | 96.66% | 84.29% | |||
| 2024-03-14 | gts | 42.45% | 27.72% | 90.57% | 77.54% | |||
| 2021-03-21 | OSKDet | 42.00% | 27.54% | 88.45% | 52.69% | |||
| 2018-11-20 | Pixel-Anchor | 40.29% | 26.10% | 88.29% | 51.88% | |||
| 2021-05-03 | NCU_MSP | 39.22% | 25.13% | 89.23% | 21.25% | |||
| 2019-03-29 | GNNets (single scale) | 38.92% | 25.45% | 82.71% | 34.04% | |||
| 2019-08-08 | JDAI | 38.52% | 24.15% | 95.21% | 77.19% | |||
| 2019-05-30 | PMTD | 38.51% | 24.22% | 93.95% | 82.23% | |||
| 2019-05-08 | Baidu-VIS | 38.13% | 24.12% | 91.00% | 22.86% | |||
| 2020-12-08 | cascade | 37.97% | 23.78% | 94.18% | 85.52% | |||
| 2021-03-25 | NCU_MSP | 37.76% | 23.92% | 89.55% | 20.06% | |||
| 2019-11-05 | baseline_maskrcnn | 37.63% | 23.40% | 95.91% | 77.70% | |||
| 2019-08-20 | juxinli | 37.55% | 23.72% | 90.06% | 50.28% | |||
| 2019-03-23 | PMTD | 37.55% | 23.71% | 90.18% | 49.86% | |||
| 2021-11-02 | fpa | 37.41% | 23.61% | 90.06% | 50.15% | |||
| 2020-09-28 | DCLNet | 37.15% | 23.61% | 87.03% | 19.73% | |||
| 2017-06-28 | SCUT_DLVClab1 | 36.60% | 23.06% | 88.68% | 72.16% | |||
| 2019-06-02 | NJU-ImagineLab | 36.43% | 22.49% | 95.80% | 82.09% | |||
| 2022-04-22 | TextBPN++(ResNet-50 with DCN) | 36.25% | 22.61% | 91.28% | 20.20% | |||
| 2019-03-19 | ccnet single scale | 35.91% | 22.58% | 87.58% | 57.28% | |||
| 2018-01-22 | FOTS_v2 | 35.83% | 22.13% | 93.95% | 71.55% | |||
| 2019-09-18 | mask RCNN Augment+ | 35.42% | 22.17% | 88.06% | 69.83% | |||
| 2018-10-29 | Amap-CVLab | 35.12% | 21.79% | 90.53% | 69.38% | |||
| 2018-11-28 | CRAFT | 35.05% | 22.27% | 82.32% | 19.53% | |||
| 2022-04-11 | TextBPN++(ResNet-50) | 34.91% | 22.07% | 83.54% | 17.60% | |||
| 2019-06-11 | 4Paradigm-Data-Intelligence | 33.95% | 20.71% | 94.15% | 20.21% | |||
| 2019-05-23 | 4Paradigm-Data-Intelligence | 33.46% | 20.43% | 92.30% | 19.04% | |||
| 2021-05-03 | adapt | 33.32% | 20.30% | 92.85% | 17.68% | |||
| 2018-05-18 | PSENet_NJU_ImagineLab (single-scale) | 33.21% | 20.94% | 80.16% | 17.24% | |||
| 2020-10-16 | Drew | 32.40% | 20.23% | 81.34% | 62.93% | |||
| 2019-07-15 | stela | 32.40% | 20.21% | 81.69% | 60.02% | |||
| 2020-12-08 | corner | 32.27% | 19.59% | 91.43% | 75.00% | |||
| 2021-12-31 | TextPMs | 31.57% | 19.52% | 82.55% | 15.31% | |||
| 2018-11-15 | USTC-NELSLIP | 31.22% | 18.74% | 93.60% | 81.67% | |||
| 2018-12-04 | SPCNet_TongJi & UESTC (multi scale) | 30.98% | 18.66% | 91.16% | 17.08% | |||
| 2021-03-03 | NCU_MSP_light | 30.96% | 18.71% | 89.59% | 15.83% | |||
| 2019-12-13 | BDN | 30.57% | 18.26% | 93.71% | 18.50% | |||
| 2023-12-17 | mlt_ch_03 | 30.02% | 18.30% | 83.38% | 14.46% | |||
| 2017-11-09 | EAST++ | 28.99% | 17.83% | 77.49% | 22.17% | |||
| 2018-08-23 | Sogou_MM | 28.94% | 17.11% | 93.60% | 78.87% | |||
| 2018-12-22 | PKU_VDIG | 28.75% | 17.01% | 92.69% | 81.18% | |||
| 2018-12-02 | Shape-Aware Based Scene Text Detector (single scale) | 28.75% | 17.26% | 86.05% | 15.68% | |||
| 2021-05-17 | NCU_FPN | 28.72% | 17.01% | 92.26% | 14.60% | |||
| 2017-06-30 | TH-DL | 28.58% | 17.37% | 80.63% | 52.72% | |||
| 2021-12-12 | a | 28.34% | 17.24% | 79.61% | 12.90% | |||
| 2020-10-21 | gccnet-ensemble | 28.14% | 16.84% | 85.54% | 53.55% | |||
| 2018-03-12 | ATL Cangjie OCR | 27.93% | 16.56% | 89.12% | 60.12% | |||
| 2021-12-12 | b | 27.79% | 16.80% | 80.51% | 12.77% | |||
| 2019-01-08 | ALGCD_CP | 27.75% | 16.50% | 87.23% | 17.27% | |||
| 2017-06-29 | SARI_FDU_RRPN_v1 | 26.38% | 15.53% | 87.39% | 61.20% | |||
| 2018-12-05 | EPTN-SJTU | 25.29% | 14.98% | 81.02% | 20.12% | |||
| 2019-05-30 | Thesis-SE | 24.04% | 14.24% | 77.13% | 14.34% | |||
| 2018-12-13 | AutoCV | 22.72% | 12.96% | 92.30% | 39.65% | |||
| 2022-01-05 | dbnet_resnet18 | 22.48% | 12.94% | 85.34% | 42.83% | |||
| 2018-12-03 | SPCNet_TongJi & UESTC (single scale) | 22.24% | 12.62% | 93.56% | 11.97% | |||
| 2017-06-28 | SARI_FDU_RRPN_v0 | 21.52% | 12.36% | 83.34% | 43.90% | |||
| 2019-01-03 | YY AI OCR Group | 16.58% | 9.56% | 62.32% | 9.54% | |||
| 2017-06-30 | Sensetime OCR | 10.32% | 5.46% | 93.44% | 60.68% | |||
| 2019-10-14 | TextSnake | 6.85% | 3.68% | 49.71% | 2.03% | |||
| 2017-06-30 | linkage-ER-Flow | 3.20% | 1.78% | 15.68% | 0.38% |