| e2e-mlt | 100.00% | 92.86% | 96.30% |
| SRC-B-TextProcessingLab | 92.31% | 100.00% | 96.00% |
| SRC-B-MachineLearningLab | 92.31% | 100.00% | 96.00% |
| HoText_v1 | 92.31% | 100.00% | 96.00% |
| Sogou_MM | 92.31% | 92.31% | 92.31% |
| Yunos_Robot1.0 | 84.62% | 100.00% | 91.67% |
| e2e-mlt + bug fix | 84.62% | 100.00% | 91.67% |
| Alibaba-PAI | 84.62% | 91.67% | 88.00% |
| Baidu IDL v3 | 76.92% | 100.00% | 86.96% |
| Sensetime V1 | 76.92% | 100.00% | 86.96% |
| FOTS | 76.92% | 100.00% | 86.96% |
 Beam search CUNI +S (decoding for TextSpotter - no postprocessing) | 76.92% | 90.91% | 83.33% |
| Baidu VIS | 76.92% | 90.91% | 83.33% |
| SRC-B-MachineLearningLab | 76.92% | 90.91% | 83.33% |
| Baidu VIS v2 | 76.92% | 90.91% | 83.33% |
| Megvii-Image++ | 76.92% | 83.33% | 80.00% |
| Baidu VIS v2+ | 76.92% | 83.33% | 80.00% |
| NVTextSpotter | 100.00% | 65.00% | 78.79% |
| TextProposals + DictNet | 76.92% | 76.92% | 76.92% |
| Tencent Youtu | 76.92% | 76.92% | 76.92% |
| CPN (multi-scale) | 76.92% | 76.92% | 76.92% |
| M_FOTS_Orig | 69.23% | 69.23% | 69.23% |
| Baidu IDL | 76.92% | 55.56% | 64.52% |
| gts-swint-cocoonly-s4-3_1 | 69.23% | 52.94% | 60.00% |
| Baseline OpenCV3.0 + Tesseract | 38.46% | 71.43% | 50.00% |
| Char-level PTSD on SynthText | 53.85% | 41.18% | 46.67% |
| Deep2Text-MO | 15.38% | 20.00% | 17.39% |
| Beam search CUNI (decoding for TextSpotter - spell checked) | 7.69% | 100.00% | 14.29% |
| Baseline-TextSpotter | 0.00% | 0.00% | 0.00% |