DeepSolo (ResNet-50)

Reported on 6 benchmarks across 1 task · 2 papers

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

Computer Vision6 results

Text SpottingonInverse-Text
F-measure (%) - Full Lexicon· 2023-05-31
53.9
best: 75.8 (DeepSolo (ViTAEv2-S, TextOCR))
DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting arXiv:2305.19957
Text SpottingonInverse-Text
F-measure (%) - No Lexicon· 2023-05-31
48.5
best: 68.8 (DeepSolo (ViTAEv2-S, TextOCR))
DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting arXiv:2305.19957
Text SpottingonSCUT-CTW1500
F-Measure (%) - Full Lexicon· 2023-05-31
81.4
best: 83.8 (SPTS)
DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting arXiv:2305.19957
Text SpottingonSCUT-CTW1500
F-measure (%) - No Lexicon· 2023-05-31
64.2
best: 64.4 (A3S)
DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting arXiv:2305.19957
Text SpottingonTotal-Text
F-measure (%) - Full Lexicon· 2022-11-19
87
best: 89.6 (DeepSolo (ViTAEv2-S, TextOCR))
DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting arXiv:2211.10772
Text SpottingonTotal-Text
F-measure (%) - No Lexicon· 2022-11-19
79.7
best: 83.6 (DeepSolo (ViTAEv2-S, TextOCR))
DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting arXiv:2211.10772