| Tencent Youtu | 100.00% | 90.91% | 95.24% |
| Baidu VIS | 80.00% | 80.00% | 80.00% |
| Baidu VIS v2 | 80.00% | 80.00% | 80.00% |
| Baidu VIS v2+ | 80.00% | 80.00% | 80.00% |
| Sogou_MM | 70.00% | 70.00% | 70.00% |
| gts-swint-cocoonly-s4-3_1 | 50.00% | 100.00% | 66.67% |
| Alibaba-PAI | 30.00% | 100.00% | 46.15% |
| SRC-B-MachineLearningLab | 30.00% | 75.00% | 42.86% |
| SRC-B-MachineLearningLab | 30.00% | 50.00% | 37.50% |
| Sensetime V1 | 20.00% | 50.00% | 28.57% |
| e2e-mlt + bug fix | 20.00% | 33.33% | 25.00% |
| Baidu IDL | 20.00% | 20.00% | 20.00% |
| TextProposals + DictNet | 10.00% | 100.00% | 18.18% |
| FOTS | 10.00% | 100.00% | 18.18% |
| M_FOTS_Orig | 10.00% | 50.00% | 16.67% |
| HoText_v1 | 10.00% | 33.33% | 15.38% |
| NVTextSpotter | 10.00% | 33.33% | 15.38% |
| Baidu IDL v3 | 10.00% | 20.00% | 13.33% |
 Baseline-TextSpotter | 0.00% | 0.00% | 0.00% |
| Beam search CUNI +S (decoding for TextSpotter - no postprocessing) | 0.00% | 0.00% | 0.00% |
| Beam search CUNI (decoding for TextSpotter - spell checked) | 0.00% | 0.00% | 0.00% |
| Deep2Text-MO | 0.00% | 0.00% | 0.00% |
| Baseline OpenCV3.0 + Tesseract | 0.00% | 0.00% | 0.00% |
| Megvii-Image++ | 0.00% | 0.00% | 0.00% |
| SRC-B-TextProcessingLab | 0.00% | 0.00% | 0.00% |
| Yunos_Robot1.0 | 0.00% | 0.00% | 0.00% |
| e2e-mlt | 0.00% | 0.00% | 0.00% |
| Char-level PTSD on SynthText | 0.00% | 0.00% | 0.00% |