| Baidu IDL v3 | 80.00% | 100.00% | 88.89% |
| SRC-B-MachineLearningLab | 80.00% | 100.00% | 88.89% |
| Tencent Youtu | 80.00% | 100.00% | 88.89% |
| Sogou_MM | 80.00% | 100.00% | 88.89% |
| HUST_MCLAB | 70.00% | 100.00% | 82.35% |
| SRC-B-MachineLearningLab | 70.00% | 100.00% | 82.35% |
| Baidu VIS | 70.00% | 100.00% | 82.35% |
| gts-swint-cocoonly-s4-3_1 | 70.00% | 100.00% | 82.35% |
| Baidu IDL | 70.00% | 87.50% | 77.78% |
| Alibaba-PAI | 70.00% | 87.50% | 77.78% |
| NVTextSpotter | 70.00% | 87.50% | 77.78% |
| Sensetime V1 | 60.00% | 100.00% | 75.00% |
| CPN (multi-scale) | 60.00% | 100.00% | 75.00% |
| e2e-mlt + bug fix | 70.00% | 77.78% | 73.68% |
| Yunos_Robot1.0 | 60.00% | 85.71% | 70.59% |
| FOTS | 60.00% | 85.71% | 70.59% |
| Smile | 60.00% | 85.71% | 70.59% |
| Baidu VIS v2+ | 70.00% | 70.00% | 70.00% |
| e2e-mlt | 70.00% | 70.00% | 70.00% |
| HoText_v1 | 60.00% | 75.00% | 66.67% |
| Baidu VIS v2 | 60.00% | 66.67% | 63.16% |
| DIY FOTS | 50.00% | 55.56% | 52.63% |
| Megvii-Image++ | 40.00% | 57.14% | 47.06% |
 NJU Text (Version3) | 30.00% | 100.00% | 46.15% |
| SRC-B-TextProcessingLab | 30.00% | 100.00% | 46.15% |
| Stradvision-2 | 30.00% | 75.00% | 42.86% |
| Beam search CUNI +S (decoding for TextSpotter - no postprocessing) | 20.00% | 100.00% | 33.33% |
| TextProposals + DictNet | 20.00% | 100.00% | 33.33% |
| Char-level PTSD on SynthText | 30.00% | 37.50% | 33.33% |
| M_FOTS_Orig | 30.00% | 33.33% | 31.58% |
| StradVision_v1 | 30.00% | 30.00% | 30.00% |
| Beam search CUNI (decoding for TextSpotter - spell checked) | 10.00% | 100.00% | 18.18% |
| Baseline-TextSpotter | 0.00% | 0.00% | 0.00% |
| Deep2Text-MO | 0.00% | 0.00% | 0.00% |
| Baseline OpenCV3.0 + Tesseract | 0.00% | 0.00% | 0.00% |