| Sensetime V1 | 75.00% | 100.00% | 85.71% |
| SRC-B-TextProcessingLab | 75.00% | 90.00% | 81.82% |
| SRC-B-MachineLearningLab | 75.00% | 90.00% | 81.82% |
| HoText_v1 | 66.67% | 100.00% | 80.00% |
| Baidu VIS v2+ | 66.67% | 100.00% | 80.00% |
| Sogou_MM | 75.00% | 75.00% | 75.00% |
| TextProposals + DictNet | 58.33% | 100.00% | 73.68% |
| FOTS | 58.33% | 100.00% | 73.68% |
| Alibaba-PAI | 58.33% | 100.00% | 73.68% |
| Yunos_Robot1.0 | 58.33% | 87.50% | 70.00% |
| Baidu VIS v2 | 58.33% | 87.50% | 70.00% |
| Tencent Youtu | 58.33% | 87.50% | 70.00% |
| SRC-B-MachineLearningLab | 50.00% | 85.71% | 63.16% |
| gts-swint-cocoonly-s4-3_1 | 50.00% | 85.71% | 63.16% |
| NVTextSpotter | 50.00% | 75.00% | 60.00% |
| Baidu VIS | 41.67% | 100.00% | 58.82% |
| Baidu IDL v3 | 41.67% | 83.33% | 55.56% |
| Baidu IDL | 50.00% | 60.00% | 54.55% |
 Beam search CUNI +S (decoding for TextSpotter - no postprocessing) | 33.33% | 66.67% | 44.44% |
| e2e-mlt + bug fix | 33.33% | 66.67% | 44.44% |
| Baseline OpenCV3.0 + Tesseract | 25.00% | 60.00% | 35.29% |
| Megvii-Image++ | 25.00% | 60.00% | 35.29% |
| Char-level PTSD on SynthText | 25.00% | 60.00% | 35.29% |
| e2e-mlt | 16.67% | 40.00% | 23.53% |
| M_FOTS_Orig | 16.67% | 20.00% | 18.18% |
| Baseline-TextSpotter | 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% |