UNITS

Reported on 5 benchmarks across 1 task · 1 paper · 3 SOTA

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

Computer Vision5 results

Text SpottingonICDAR 2015
F-measure (%) - Generic Lexicon· 2023-04-07
80.3
SOTA
Towards Unified Scene Text Spotting based on Sequence Generation arXiv:2304.03435
Text SpottingonICDAR 2015
F-measure (%) - Strong Lexicon· 2023-04-07
89
SOTA
Towards Unified Scene Text Spotting based on Sequence Generation arXiv:2304.03435
Text SpottingonICDAR 2015
F-measure (%) - Weak Lexicon· 2023-04-07
84.1
SOTA
Towards Unified Scene Text Spotting based on Sequence Generation arXiv:2304.03435
Text SpottingonTotal-Text
F-measure (%) - Full Lexicon· uses extra data· 2023-04-07
86
best: 89.6 (DeepSolo (ViTAEv2-S, TextOCR))
Towards Unified Scene Text Spotting based on Sequence Generation arXiv:2304.03435
Text SpottingonTotal-Text
F-measure (%) - No Lexicon· uses extra data· 2023-04-07
78.7
best: 83.6 (DeepSolo (ViTAEv2-S, TextOCR))
Towards Unified Scene Text Spotting based on Sequence Generation arXiv:2304.03435