Scene Text Recognition on WOST
91.2Word AccuracyCSD-B
Evaluation Results
| Method | Links | |
|---|---|---|
| CSD-BData=RBU, Params=104.9 M2025.03 | 91.2 | |
| CSD-SData=Real, Params=40.8 M2025.03 | 90.9 | |
| CLIP4STR-HData=RBU, Params=1 B2025.03 | 90.9 | |
| CSD-DData=Real, Params=110.2 M2025.03 | 90.8 | |
| CLIP4STR-LData=RBU, Params=446 M2025.03 | 90.6 | |
| CSD-DData=RBU, Params=110.2 M2025.03 | 90.6 | |
| CSD-BData=Real, Params=104.9 M2025.03 | 90 | |
| CSD-SData=RBU, Params=40.8 M2025.03 | 89.5 | |
| CLIP4STR-LData=Real, Params=446 M2025.03 | 89.2 | |
| CLIP4STR-BData=RBU, Params=158 M2025.03 | 88.8 | |
| CLIP4STR-BData=Real, Params=158 M2025.03 | 87.5 | |
| PARSeqData=Real, Params=22.5 M2025.03 | 85.4 | |
| ABINETData=Real, Params=23.5 M2025.03 | 85 | |
| VITSTR-SData=Real, Params=21.7 M2025.03 | 77.9 |