Sample RankingRecallPrecisionHmean
Alibaba-PAI84.62%84.62%84.62%
gts-swint-cocoonly-s4-3_184.62%84.62%84.62%
CPN (multi-scale)84.62%84.62%84.62%
Sogou_MM76.92%90.91%83.33%
Baidu IDL76.92%76.92%76.92%
Baidu IDL v376.92%76.92%76.92%
Sensetime V176.92%76.92%76.92%
HoText_v176.92%76.92%76.92%
Baidu VIS v2+76.92%76.92%76.92%
Tencent Youtu69.23%81.82%75.00%
Smile76.92%71.43%74.07%
FOTS61.54%80.00%69.57%
Baidu VIS69.23%69.23%69.23%
Baidu VIS v269.23%69.23%69.23%
NVTextSpotter69.23%69.23%69.23%
e2e-mlt + bug fix61.54%61.54%61.54%
SRC-B-MachineLearningLab61.54%57.14%59.26%
SRC-B-MachineLearningLab61.54%57.14%59.26%
e2e-mlt53.85%58.33%56.00%
TextProposals + DictNet38.46%100.00%55.56%
M_FOTS_Orig53.85%53.85%53.85%
Yunos_Robot1.038.46%62.50%47.62%
Megvii-Image++38.46%45.45%41.67%
Char-level PTSD on SynthText23.08%42.86%30.00%
Beam search CUNI +S (decoding for TextSpotter - no postprocessing)15.38%40.00%22.22%
SRC-B-TextProcessingLab15.38%28.57%20.00%
Deep2Text-MO7.69%25.00%11.76%
Baseline-TextSpotter0.00%0.00%0.00%
Beam search CUNI (decoding for TextSpotter - spell checked)0.00%0.00%0.00%
Baseline OpenCV3.0 + Tesseract0.00%0.00%0.00%