| LRANet | 100.00% | 82.61% | 90.48% |
 baseline_polygon | 94.74% | 78.26% | 85.71% |
| I3CL | 89.47% | 80.95% | 85.00% |
| AntFin-Cascade Mask R-CNN | 84.21% | 76.19% | 80.00% |
| NJU-ImagineLab | 78.95% | 78.95% | 78.95% |
| DuXiaoman_OCR | 78.95% | 75.00% | 76.92% |
| Fudan-Supremind Detection v3 | 84.21% | 69.57% | 76.19% |
| TMIS | 68.42% | 81.25% | 74.29% |
| Tencent TEG OCR | 73.68% | 73.68% | 73.68% |
| TextFuseNet (ResNeXt-101) | 73.68% | 73.68% | 73.68% |
| HOCRA | 63.16% | 85.71% | 72.73% |
| Sg_ptd | 63.16% | 80.00% | 70.59% |
| TP | 68.42% | 72.22% | 70.27% |
| CUTeOCR | 73.68% | 66.67% | 70.00% |
| NN_Chinese_and_euro6 | 73.68% | 66.67% | 70.00% |
| A scene text detection method based on maskrcnn | 68.42% | 65.00% | 66.67% |
| AutoCV | 68.42% | 65.00% | 66.67% |
| TD-PPIoU (Long-Pretrain) | 52.63% | 90.91% | 66.67% |
| TextFuseNet (ResNet-50) | 63.16% | 66.67% | 64.86% |
| QAQ | 57.89% | 73.33% | 64.71% |
| mask rcnn | 68.42% | 59.09% | 63.41% |
| SRCB_Art | 63.16% | 63.16% | 63.16% |
| CCISTD | 63.16% | 63.16% | 63.16% |
| Unicamp-SRBR-PN-1 | 63.16% | 63.16% | 63.16% |
| MaskRCNN_Text | 57.89% | 68.75% | 62.86% |
| NN_euro6 | 52.63% | 76.92% | 62.50% |
| DPText-DETR (ResNet-50) | 52.63% | 76.92% | 62.50% |
| fdu_ai | 57.89% | 64.71% | 61.11% |
| CPNText-DETR(resnet-50) | 52.63% | 71.43% | 60.61% |
| ArtDet-v2 | 57.89% | 61.11% | 59.46% |
| DMText_art | 42.11% | 100.00% | 59.26% |
| Art detect by vivo | 52.63% | 66.67% | 58.82% |
| Alibaba-PAI | 63.16% | 54.55% | 58.54% |
| SRFormer (ResNet50-#1seg) | 52.63% | 62.50% | 57.14% |
| Art_test_baseline_task1 | 63.16% | 50.00% | 55.81% |
| PAT-S.Y | 52.63% | 58.82% | 55.56% |
| Sogou_OCR | 52.63% | 55.56% | 54.05% |
| TEXT_SNIPER | 52.63% | 52.63% | 52.63% |
| MFTD: Mask Filters for Text Detection | 57.89% | 47.83% | 52.38% |
| Mask RCNN | 68.42% | 41.94% | 52.00% |
| HOCRA_base | 36.84% | 87.50% | 51.85% |
| TD-PPIoU | 42.11% | 66.67% | 51.61% |
| CLTDR | 47.37% | 56.25% | 51.43% |
| MEGVII_Detection | 36.84% | 77.78% | 50.00% |
| TextBPN++(ResNet-50 with DCN) | 42.11% | 61.54% | 50.00% |
| I3CL(ViTAEv2-S) | 42.11% | 61.54% | 50.00% |
| MaskDet | 52.63% | 43.48% | 47.62% |
| TextBPN++(ResNet-50) | 42.11% | 53.33% | 47.06% |
| DMCA | 52.63% | 41.67% | 46.51% |
| Improved Progressive scale expansion Net | 47.37% | 45.00% | 46.15% |
| TextMask_V1 | 47.37% | 45.00% | 46.15% |
| CRAFT | 42.11% | 44.44% | 43.24% |
| EM-DATA | 42.11% | 44.44% | 43.24% |
| 1 | 52.63% | 35.71% | 42.55% |
| ESRNet | 36.84% | 43.75% | 40.00% |
| Mask R-CNN | 26.32% | 38.46% | 31.25% |
| dp_pq_nn | 21.05% | 57.14% | 30.77% |
| TextCohesion_1 | 31.58% | 23.08% | 26.67% |
| dp_nn | 21.05% | 33.33% | 25.81% |
| inception baseline | 21.05% | 30.77% | 25.00% |
| RAST: Robust Arbitrary Shape Text Detector | 10.53% | 13.33% | 11.76% |
| MSR | 0.00% | 0.00% | 0.00% |