Sample RankingRecallPrecisionHmean
HoText_v175.00%75.00%75.00%
TextProposals + DictNet58.33%100.00%73.68%
Yunos_Robot1.058.33%87.50%70.00%
Baidu IDL66.67%72.73%69.57%
Sogou_MM66.67%72.73%69.57%
Tencent Youtu50.00%85.71%63.16%
Baidu VIS v2+50.00%85.71%63.16%
gts-swint-cocoonly-s4-3_150.00%85.71%63.16%
SRC-B-TextProcessingLab58.33%63.64%60.87%
NVTextSpotter50.00%75.00%60.00%
Smile50.00%66.67%57.14%
FOTS41.67%83.33%55.56%
SRC-B-MachineLearningLab41.67%71.43%52.63%
SRC-B-MachineLearningLab41.67%62.50%50.00%
Baidu VIS v241.67%62.50%50.00%
Alibaba-PAI41.67%50.00%45.45%
Beam search CUNI +S (decoding for TextSpotter - no postprocessing)33.33%66.67%44.44%
Baidu VIS33.33%66.67%44.44%
e2e-mlt + bug fix33.33%66.67%44.44%
Baidu IDL v333.33%57.14%42.11%
Char-level PTSD on SynthText25.00%60.00%35.29%
Sensetime V133.33%36.36%34.78%
Baseline OpenCV3.0 + Tesseract16.67%40.00%23.53%
Megvii-Image++16.67%40.00%23.53%
e2e-mlt16.67%40.00%23.53%
M_FOTS_Orig16.67%20.00%18.18%
Baseline-TextSpotter0.00%0.00%0.00%
Beam search CUNI (decoding for TextSpotter - spell checked)0.00%0.00%0.00%
Deep2Text-MO0.00%0.00%0.00%