MaskTextSpotter v3

Reported on 5 benchmarks across 1 task · 1 paper · 4 SOTA

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

Computer Vision5 results

Text SpottingonTotal-Text
F-measure (%) - Full Lexicon· 2020-07-18
78.4
best: 89.6 (DeepSolo (ViTAEv2-S, TextOCR))
SOTA
Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting arXiv:2007.09482
Text SpottingonTotal-Text
F-measure (%) - No Lexicon· 2020-07-18
71.2
best: 83.6 (DeepSolo (ViTAEv2-S, TextOCR))
SOTA
Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting arXiv:2007.09482
Text SpottingonICDAR 2015
F-measure (%) - Generic Lexicon· 2020-07-18
74.2
best: 80.3 (UNITS)
SOTA
Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting arXiv:2007.09482
Text SpottingonICDAR 2015
F-measure (%) - Weak Lexicon· 2020-07-18
78.1
best: 84.1 (UNITS)
SOTA
Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting arXiv:2007.09482
Text SpottingonICDAR 2015
F-measure (%) - Strong Lexicon· 2020-07-18
83.3
best: 89 (UNITS)
Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting arXiv:2007.09482