| Baidu IDL | 69.23% | 75.00% | 72.00% |
| HoText_v1 | 69.23% | 75.00% | 72.00% |
| Sogou_MM | 69.23% | 75.00% | 72.00% |
| TextProposals + DictNet | 53.85% | 100.00% | 70.00% |
| Yunos_Robot1.0 | 53.85% | 87.50% | 66.67% |
| gts-swint-cocoonly-s4-3_1 | 53.85% | 87.50% | 66.67% |
| NVTextSpotter | 53.85% | 77.78% | 63.64% |
| FOTS | 46.15% | 85.71% | 60.00% |
| Tencent Youtu | 46.15% | 85.71% | 60.00% |
| Baidu VIS v2+ | 46.15% | 85.71% | 60.00% |
| SRC-B-TextProcessingLab | 53.85% | 63.64% | 58.33% |
| Smile | 46.15% | 60.00% | 52.17% |
| SRC-B-MachineLearningLab | 38.46% | 71.43% | 50.00% |
| Alibaba-PAI | 46.15% | 54.55% | 50.00% |
| Baidu IDL v3 | 38.46% | 62.50% | 47.62% |
| SRC-B-MachineLearningLab | 38.46% | 62.50% | 47.62% |
| Baidu VIS v2 | 38.46% | 62.50% | 47.62% |
 Beam search CUNI +S (decoding for TextSpotter - no postprocessing) | 30.77% | 66.67% | 42.11% |
| Baidu VIS | 30.77% | 66.67% | 42.11% |
| e2e-mlt + bug fix | 30.77% | 66.67% | 42.11% |
| Sensetime V1 | 30.77% | 36.36% | 33.33% |
| Char-level PTSD on SynthText | 23.08% | 60.00% | 33.33% |
| Baseline OpenCV3.0 + Tesseract | 15.38% | 40.00% | 22.22% |
| Megvii-Image++ | 15.38% | 40.00% | 22.22% |
| e2e-mlt | 15.38% | 40.00% | 22.22% |
| M_FOTS_Orig | 15.38% | 20.00% | 17.39% |
| 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% |