Sample RankingRecallPrecisionHmean
e2e-mlt + bug fix83.33%100.00%90.91%
SRC-B-MachineLearningLab83.33%83.33%83.33%
M_FOTS_Orig83.33%83.33%83.33%
TextProposals + DictNet66.67%80.00%72.73%
SRC-B-MachineLearningLab66.67%80.00%72.73%
Baidu IDL v366.67%80.00%72.73%
Sensetime V166.67%80.00%72.73%
e2e-mlt66.67%80.00%72.73%
Sogou_MM66.67%80.00%72.73%
Alibaba-PAI66.67%80.00%72.73%
HoText_v166.67%66.67%66.67%
NVTextSpotter50.00%75.00%60.00%
SRC-B-TextProcessingLab50.00%60.00%54.55%
FOTS33.33%66.67%44.44%
Yunos_Robot1.033.33%50.00%40.00%
Baidu VIS33.33%50.00%40.00%
Beam search CUNI +S (decoding for TextSpotter - no postprocessing)33.33%40.00%36.36%
Baidu IDL33.33%40.00%36.36%
Baidu VIS v233.33%40.00%36.36%
Tencent Youtu33.33%40.00%36.36%
gts-swint-cocoonly-s4-3_133.33%40.00%36.36%
Baseline OpenCV3.0 + Tesseract16.67%100.00%28.57%
Smile16.67%25.00%20.00%
Baidu VIS v2+16.67%25.00%20.00%
Deep2Text-MO16.67%14.29%15.38%
Baseline-TextSpotter0.00%0.00%0.00%
Beam search CUNI (decoding for TextSpotter - spell checked)0.00%0.00%0.00%
Megvii-Image++0.00%0.00%0.00%
Char-level PTSD on SynthText0.00%0.00%0.00%