WordSup (VGG16-synth-icdar)

Reported on 5 benchmarks across 1 task · 2 papers · 5 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision5 results

Scene Text DetectiononICDAR 2013
Precision· 2017-08-22
93.34
best: 97.4 (CRAFT)
SOTA
WordSup: Exploiting Word Annotations for Character based Text Detection arXiv:1708.06720
Scene Text DetectiononICDAR 2013
Recall· 2017-08-22
87.53
best: 93.1 (CRAFT)
SOTA
WordSup: Exploiting Word Annotations for Character based Text Detection arXiv:1708.06720
Scene Text DetectiononICDAR 2015
F-Measure· 2017-03-19
78.2
best: 92.23 (TextFuseNet (ResNeXt-101))
SOTA
Detecting Oriented Text in Natural Images by Linking Segments arXiv:1703.06520
Scene Text DetectiononICDAR 2015
Precision· 2017-03-19
79.3
best: 93.96 (TextFuseNet (ResNeXt-101))
SOTA
Detecting Oriented Text in Natural Images by Linking Segments arXiv:1703.06520
Scene Text DetectiononICDAR 2015
Recall· 2017-03-19
77
best: 91.98 (CharNet H-88 (single-scale))
SOTA
Detecting Oriented Text in Natural Images by Linking Segments arXiv:1703.06520