Sample RankingRecallPrecisionHmean
Baidu IDL v3100.00%100.00%100.00%
Sensetime V175.00%100.00%85.71%
FOTS75.00%100.00%85.71%
Alibaba-PAI75.00%100.00%85.71%
gts-swint-cocoonly-s4-3_175.00%100.00%85.71%
Baidu VIS v275.00%75.00%75.00%
Tencent Youtu75.00%75.00%75.00%
Baidu VIS v2+75.00%75.00%75.00%
Sogou_MM75.00%75.00%75.00%
Baidu VIS50.00%100.00%66.67%
SRC-B-MachineLearningLab50.00%100.00%66.67%
CPN (multi-scale)50.00%100.00%66.67%
Yunos_Robot1.050.00%66.67%57.14%
Baidu IDL50.00%50.00%50.00%
HoText_v150.00%50.00%50.00%
e2e-mlt50.00%50.00%50.00%
Deep2Text-MO25.00%100.00%40.00%
Baseline OpenCV3.0 + Tesseract25.00%100.00%40.00%
TextProposals + DictNet25.00%100.00%40.00%
SRC-B-TextProcessingLab25.00%100.00%40.00%
Megvii-Image++25.00%50.00%33.33%
SRC-B-MachineLearningLab25.00%50.00%33.33%
e2e-mlt + bug fix25.00%33.33%28.57%
Char-level PTSD on SynthText25.00%33.33%28.57%
NVTextSpotter25.00%33.33%28.57%
M_FOTS_Orig25.00%25.00%25.00%
Baseline-TextSpotter0.00%0.00%0.00%
Beam search CUNI +S (decoding for TextSpotter - no postprocessing)0.00%0.00%0.00%
Beam search CUNI (decoding for TextSpotter - spell checked)0.00%0.00%0.00%