Sample RankingRecallPrecisionHmean
Alibaba-PAI100.00%87.50%93.33%
Sensetime V185.71%100.00%92.31%
FOTS85.71%100.00%92.31%
Smile85.71%100.00%92.31%
Sogou_MM85.71%100.00%92.31%
NVTextSpotter85.71%100.00%92.31%
Baidu IDL85.71%85.71%85.71%
Baidu IDL v385.71%85.71%85.71%
Baidu VIS v285.71%85.71%85.71%
Tencent Youtu85.71%85.71%85.71%
Baidu VIS v2+85.71%85.71%85.71%
HoText_v171.43%100.00%83.33%
SRC-B-MachineLearningLab71.43%100.00%83.33%
Char-level PTSD on SynthText71.43%100.00%83.33%
SRC-B-MachineLearningLab71.43%83.33%76.92%
Baidu VIS71.43%83.33%76.92%
gts-swint-cocoonly-s4-3_171.43%83.33%76.92%
CPN (multi-scale)71.43%83.33%76.92%
DIY FOTS42.86%100.00%60.00%
e2e-mlt57.14%57.14%57.14%
e2e-mlt + bug fix57.14%57.14%57.14%
TextProposals + DictNet42.86%75.00%54.55%
M_FOTS_Orig42.86%75.00%54.55%
SRC-B-TextProcessingLab28.57%100.00%44.44%
Yunos_Robot1.028.57%66.67%40.00%
HUST_MCLAB28.57%66.67%40.00%
NJU Text (Version3)28.57%50.00%36.36%
Beam search CUNI +S (decoding for TextSpotter - no postprocessing)14.29%100.00%25.00%
Stradvision-214.29%100.00%25.00%
Deep2Text-MO14.29%50.00%22.22%
Baseline OpenCV3.0 + Tesseract14.29%33.33%20.00%
Megvii-Image++14.29%33.33%20.00%
StradVision_v114.29%25.00%18.18%
Baseline-TextSpotter0.00%0.00%0.00%
Beam search CUNI (decoding for TextSpotter - spell checked)0.00%0.00%0.00%