| AntFin-Cascade Mask R-CNN | 95.12% | 100.00% | 97.50% |
 MEGVII_Detection | 92.68% | 97.44% | 95.00% |
| Mask R-CNN | 95.12% | 88.64% | 91.76% |
| I3CL | 90.24% | 92.50% | 91.36% |
| DuXiaoman_OCR | 85.37% | 87.50% | 86.42% |
| CPNText-DETR(resnet-50) | 85.37% | 87.50% | 86.42% |
| SRFormer (ResNet50-#1seg) | 82.93% | 89.47% | 86.08% |
| DPText-DETR (ResNet-50) | 85.37% | 85.37% | 85.37% |
| I3CL(ViTAEv2-S) | 80.49% | 89.19% | 84.62% |
| HOCRA | 92.68% | 77.55% | 84.44% |
| ESRNet | 85.37% | 81.40% | 83.33% |
| TD-PPIoU (Long-Pretrain) | 78.05% | 88.89% | 83.12% |
| TD-PPIoU | 80.49% | 82.50% | 81.48% |
| TextFuseNet (ResNeXt-101) | 75.61% | 83.78% | 79.49% |
| Sg_ptd | 70.73% | 90.63% | 79.45% |
| TextFuseNet (ResNet-50) | 70.73% | 87.88% | 78.38% |
| AutoCV | 73.17% | 83.33% | 77.92% |
| TextBPN++(ResNet-50 with DCN) | 73.17% | 83.33% | 77.92% |
| HOCRA_base | 63.41% | 100.00% | 77.61% |
| ArtDet-v2 | 70.73% | 85.29% | 77.33% |
| baseline_polygon | 73.17% | 81.08% | 76.92% |
| LRANet | 73.17% | 78.95% | 75.95% |
| Tencent TEG OCR | 68.29% | 84.85% | 75.68% |
| CUTeOCR | 65.85% | 87.10% | 75.00% |
| Sogou_OCR | 68.29% | 75.68% | 71.79% |
| Alibaba-PAI | 63.41% | 76.47% | 69.33% |
| SRCB_Art | 63.41% | 76.47% | 69.33% |
| CCISTD | 53.66% | 84.62% | 65.67% |
| TextBPN++(ResNet-50) | 73.17% | 58.82% | 65.22% |
| NJU-ImagineLab | 56.10% | 76.67% | 64.79% |
| MaskDet | 56.10% | 74.19% | 63.89% |
| Unicamp-SRBR-PN-1 | 58.54% | 68.57% | 63.16% |
| DMText_art | 56.10% | 69.70% | 62.16% |
| A scene text detection method based on maskrcnn | 51.22% | 77.78% | 61.76% |
| NN_euro6 | 51.22% | 75.00% | 60.87% |
| TextMask_V1 | 53.66% | 66.67% | 59.46% |
| dp_nn | 53.66% | 66.67% | 59.46% |
| dp_pq_nn | 53.66% | 66.67% | 59.46% |
| MFTD: Mask Filters for Text Detection | 56.10% | 60.53% | 58.23% |
| CRAFT | 56.10% | 60.53% | 58.23% |
| TMIS | 48.78% | 68.97% | 57.14% |
| Mask RCNN | 56.10% | 57.50% | 56.79% |
| TEXT_SNIPER | 48.78% | 66.67% | 56.34% |
| NN_Chinese_and_euro6 | 48.78% | 66.67% | 56.34% |
| mask rcnn | 43.90% | 75.00% | 55.38% |
| Art detect by vivo | 43.90% | 66.67% | 52.94% |
| MaskRCNN_Text | 46.34% | 61.29% | 52.78% |
| Fudan-Supremind Detection v3 | 48.78% | 57.14% | 52.63% |
| Art_test_baseline_task1 | 43.90% | 54.55% | 48.65% |
| Improved Progressive scale expansion Net | 39.02% | 64.00% | 48.48% |
| TextCohesion_1 | 34.15% | 66.67% | 45.16% |
| fdu_ai | 34.15% | 60.87% | 43.75% |
| DMCA | 39.02% | 48.48% | 43.24% |
| TP | 34.15% | 56.00% | 42.42% |
| QAQ | 29.27% | 41.38% | 34.29% |
| CLTDR | 24.39% | 35.71% | 28.99% |
| 1 | 21.95% | 33.33% | 26.47% |
| EM-DATA | 14.63% | 27.27% | 19.05% |
| RAST: Robust Arbitrary Shape Text Detector | 12.20% | 27.78% | 16.95% |
| PAT-S.Y | 4.88% | 50.00% | 8.89% |
| inception baseline | 4.88% | 15.38% | 7.41% |
| MSR | 0.00% | 0.00% | 0.00% |