Scene Text Recognition on ArT 34k samples
86.4Word AccuracyCLIP4STR-L
Evaluation Results
| Method | Links | |
|---|---|---|
| CLIP4STR-LData=RBU, Params=446 M2025.03 | 86.4 | |
| CLIP4STR-HData=RBU, Params=1 B2025.03 | 86.4 | |
| CSD-BData=RBU, Params=104.9 M2025.03 | 86.4 | |
| CSD-DData=RBU, Params=110.2 M2025.03 | 86.4 | |
| CSD-SData=RBU, Params=40.8 M2025.03 | 86.3 | |
| CSD-SData=Real, Params=40.8 M2025.03 | 86.2 | |
| CSD-BData=Real, Params=104.9 M2025.03 | 86.2 | |
| CSD-DData=Real, Params=110.2 M2025.03 | 86.1 | |
| CLIP4STR-LData=Real, Params=446 M2025.03 | 85.9 | |
| CLIP4STR-BData=Real, Params=158 M2025.03 | 85.8 | |
| CLIP4STR-BData=RBU, Params=158 M2025.03 | 85.8 | |
| PARSeqData=Real, Params=22.5 M2025.03 | 84.5 | |
| TRBAData=Real, Params=49.6 M2025.03 | 82.5 | |
| ABINETData=Real, Params=23.5 M2025.03 | 81.2 | |
| VITSTR-SData=Real, Params=21.7 M2025.03 | 81.1 | |
| CRNNData=Real, Params=8.5 M2025.03 | 66.8 |