Scene Text Recognition on ArT (test)
86.4Word AccuracyCLIP4STR-L
Evaluation Results
| Method | Links | |
|---|---|---|
| CLIP4STR-LTrain data=RBU(6.5M)2023.05 | 86.4 | |
| CLIP4STR-HTrain data=RBU(6.5M)2023.05 | 86.4 | |
| CLIP4STR-LTrain data=Real(3.3M)2023.05 | 85.9 | |
| CLIP4STR-BTrain data=Real(3.3M)2023.05 | 85.8 | |
| CLIP4STR-BTrain data=RBU(6.5M)2023.05 | 85.8 | |
| PARSeqTrain data=Real(3.3M)2023.05 | 84.5 | |
| MPSTRTrain data=Real(3.3M)2023.05 | 84.4 | |
| ViTSTR-STrain data=Real(3.3M)2023.05 | 82.5 | |
| TRBATrain data=Real(3.3M)2023.05 | 82.5 | |
| ABINetTrain data=Real(3.3M)2023.05 | 81.2 | |
| DiG-ViT-BTrain data=Real(2.8M)2023.05 | 81 | |
| CLIP4STR-LTrain data=MJ+ST2023.05 | 73.7 | |
| CLIP4STR-BTrain data=MJ+ST2023.05 | 72.8 | |
| PARSeqTrain data=MJ+ST2023.05 | 70.7 | |
| CLIP-OCRTrain data=MJ+ST2023.05 | 70.5 | |
| MPSTRTrain data=MJ+ST2023.05 | 69.9 | |
| TRBATrain data=MJ+ST2023.05 | 68.2 | |
| ViTSTR-STrain data=MJ+ST2023.05 | 66.1 | |
| ABINetTrain data=MJ+ST2023.05 | 65.4 |