- Task 1 - Text Localization
- Task 2 - Script identification
- Task 3 - Joint text detection and script identification
- Task 4 - End-to-End text detection and recognition
method: TH2020-04-19
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_OCR2019-11-11
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: Tencent-DPPR Team2019-06-04
Authors: Longhuang Wu, Shangxuan Tian, Chang Liu, Wenjie Cai, Jiachen Li, Sicong Liu, Haoxi Li, Chunchao Guo, Hongfa Wang, Hongkai Chen, Qinglin lu, Chun Yang, Xucheng Yin, Lei Xiao
Description: We are Tencent-DPPR (Data Platform Precision Recommendation) team. Our method follows the framework of Mask R-CNN that employs mask to detect multi-oriented scene texts. We use the MLT-19 and the MSRA-TD500 dataset to train our text detector, and we also apply a multi-scale training approach during training. To obtain the final ensemble results, we combined two different backbones and different multi-scale testing approaches.
| Date | Method | Hmean | Precision | Recall | Average Precision | |||
|---|---|---|---|---|---|---|---|---|
| 2020-04-19 | TH | 84.09% | 86.67% | 81.65% | 78.40% | |||
| 2019-11-11 | Sogou_OCR | 82.99% | 86.15% | 80.05% | 76.55% | |||
| 2019-06-04 | Tencent-DPPR Team | 82.75% | 84.26% | 81.30% | 77.56% | |||
| 2019-06-04 | Tencent-DPPR Team (Method_v0.3) | 82.75% | 84.29% | 81.25% | 77.51% | |||
| 2019-06-03 | Tencent-DPPR Team (Method_v0.2) | 82.69% | 84.50% | 80.96% | 77.25% | |||
| 2019-06-04 | multi-stage_text_detector_v4 | 82.65% | 84.62% | 80.76% | 68.43% | |||
| 2019-06-03 | NJU-ImagineLab(v3) | 82.42% | 84.06% | 80.84% | 77.21% | |||
| 2019-06-03 | multi-stage_text_detector | 82.37% | 84.39% | 80.45% | 68.04% | |||
| 2019-06-04 | multi-stage_text_detector_v3 | 82.33% | 84.04% | 80.69% | 67.96% | |||
| 2019-06-04 | multi-stage_text_detector_v2 | 82.29% | 83.91% | 80.74% | 67.90% | |||
| 2019-05-30 | PMTD | 81.79% | 83.96% | 79.74% | 76.65% | |||
| 2019-05-27 | Tencent-DPPR Team (Method_v0.1) | 81.21% | 86.65% | 76.41% | 72.62% | |||
| 2021-03-11 | SituTech_OCR | 80.51% | 89.31% | 73.30% | 65.35% | |||
| 2019-05-29 | IC_RL | 79.19% | 79.06% | 79.32% | 62.97% | |||
| 2019-05-29 | maskrcnn++ result | 79.03% | 78.32% | 79.75% | 62.74% | |||
| 2019-06-02 | A two-stage text detector based on cascade rcnn(using total 10000 images of mlt19) | 78.21% | 79.31% | 77.13% | 72.54% | |||
| 2021-02-04 | NCU_MSP | 78.09% | 80.42% | 75.90% | 60.81% | |||
| 2019-05-31 | A two-stage text detector based on cascade rcnn | 77.90% | 79.94% | 75.96% | 71.42% | |||
| 2022-11-02 | ESTextSpotter | 77.34% | 79.33% | 75.45% | 71.17% | |||
| 2019-05-27 | TH-DL | 76.78% | 83.33% | 71.19% | 65.06% | |||
| 2019-06-04 | TH-DL-v2 | 76.70% | 82.36% | 71.76% | 65.23% | |||
| 2019-06-03 | TH-DL-v1 | 76.59% | 82.34% | 71.59% | 65.14% | |||
| 2019-06-03 | mm-maskrcnn_v2 | 75.86% | 81.49% | 70.96% | 67.15% | |||
| 2020-10-16 | Drew | 75.71% | 81.06% | 71.02% | 66.56% | |||
| 2019-06-02 | DISTILLED CRAFT | 75.61% | 81.81% | 70.29% | 63.82% | |||
| 2023-05-22 | DeepSolo++ (ResNet-50) | 74.93% | 82.89% | 68.36% | 65.63% | |||
| 2020-05-30 | NCU | 74.29% | 73.90% | 74.68% | 54.64% | |||
| 2019-05-26 | two stage text detector | 74.08% | 78.70% | 69.97% | 65.38% | |||
| 2019-06-03 | CRAFTS | 72.49% | 80.63% | 65.84% | 59.84% | |||
| 2019-06-03 | sot | 72.10% | 75.57% | 68.93% | 64.28% | |||
| 2023-05-30 | TD-PPIoU | 71.10% | 68.45% | 73.97% | 68.66% | |||
| 2019-06-03 | text-mountain | 71.02% | 69.67% | 72.43% | 50.95% | |||
| 2019-06-04 | Unicamp-SRBR-MLT2019-PELEETEXT | 68.25% | 76.04% | 61.91% | 56.83% | |||
| 2019-06-03 | RRPN | 68.01% | 73.62% | 63.19% | 57.02% | |||
| 2019-05-28 | CRAFTS(Initial) | 67.88% | 76.95% | 60.72% | 56.01% | |||
| 2019-06-04 | Unicamp-SRBR-MLT2019-FUSION-PSENET-PELEETEXT | 65.87% | 72.57% | 60.31% | 53.05% | |||
| 2019-06-04 | Lomin OCR | 65.84% | 67.23% | 64.51% | 56.70% | |||
| 2019-05-24 | PSENet_v1 | 65.77% | 73.21% | 59.69% | 52.45% | |||
| 2019-06-03 | NXB OCR | 63.38% | 64.04% | 62.73% | 40.13% | |||
| 2019-05-27 | CLTDR | 61.73% | 73.94% | 52.98% | 39.35% | |||
| 2019-05-27 | MLT2019 ETD | 60.43% | 72.45% | 51.83% | 37.73% | |||
| 2020-10-07 | MEAST_V2_8_oct | 59.72% | 66.15% | 54.44% | 38.78% | |||
| 2020-10-23 | MEAST_V3_23_Oct | 59.62% | 64.91% | 55.13% | 38.74% | |||
| 2019-05-27 | NXB OCR | 59.26% | 67.52% | 52.81% | 36.99% | |||
| 2019-06-03 | TP | 54.98% | 74.40% | 43.61% | 34.10% | |||
| 2019-05-28 | Unicamp-SRBR-MLT2019-S1 | 46.05% | 71.89% | 33.88% | 30.19% | |||
| 2019-06-04 | Cyberspace | 42.47% | 58.18% | 33.44% | 21.05% | |||
| 2019-05-28 | PydBox-TextDetector | 35.92% | 66.64% | 24.58% | 16.63% | |||
| 2020-12-15 | DSIT-UOA | 21.12% | 21.03% | 21.20% | 5.85% | |||
| 2019-05-05 | AAAA | 0.02% | 0.03% | 0.01% | 0.00% | |||
| 2019-05-27 | 4Paradigm-Data-Intelligence | 0.00% | 0.00% | 0.00% | 0.00% | |||
| 2019-05-27 | Unicamp-SRBR-MLT2019-S1 | 0.00% | 0.00% | 0.00% | 0.00% | |||
| 2019-06-01 | tsinghuaee51_MLT2019 | 0.00% | 0.00% | 0.00% | 0.00% |