Text Spotting On Inverse Text
評価指標
F-measure (%) - Full Lexicon
F-measure (%) - No Lexicon
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | F-measure (%) - Full Lexicon | F-measure (%) - No Lexicon | Paper Title | Repository |
|---|---|---|---|---|
| ABCNet v2 | 47.4 | 34.5 | ABCNet v2: Adaptive Bezier-Curve Network for Real-time End-to-end Text Spotting | |
| DeepSolo (ResNet-50) | 53.9 | 48.5 | DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting | |
| SwinTextSpotter | 67.9 | 55.4 | SwinTextSpotter: Scene Text Spotting via Better Synergy between Text Detection and Text Recognition | |
| DeepSolo (ResNet-50, TextOCR) | 71.2 | 64.6 | DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting | |
| TESTR | 41.6 | 34.2 | Text Spotting Transformers | |
| SPTS | 46.2 | 38.3 | SPTS: Single-Point Text Spotting | |
| MaskTextSpotter v2 | 43.5 | 39.0 | Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes | |
| ABCNet | 34.3 | 22.2 | ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network | |
| DeepSolo (ViTAEv2-S, TextOCR) | 75.8 | 68.8 | DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting |
0 of 9 row(s) selected.