Text Recognition on BCTR
73.9Accuracy (Scene)MaskOCR (ViT-B)
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MaskOCR (ViT-B)Visual Pre-training=true, Language Pre-training=true, Parameters=100M2022.06 | 73.9 | 74.8 | 99.3 | 63.7 | 80.8 | |
| MaskOCR (ViT-S)Visual Pre-training=true, Language Pre-training=true, Parameters=36M2022.06 | 71.4 | 72.5 | 98.8 | 55.6 | 78.1 | |
| MaskOCR (ViT-B)Visual Pre-training=false, Language Pre-training=false, Parameters=100M2022.06 | 68.8 | 70.7 | 98.6 | 49.4 | 75.8 | |
| MaskOCR (ViT-S)Visual Pre-training=false, Language Pre-training=false, Parameters=36M2022.06 | 68.6 | 70.3 | 98.5 | 49.1 | 75.6 | |
| TransOCRVisual Pre-training=false, Language Pre-training=false, Parameters=84M2022.06 | 63.3 | 62.3 | 96.9 | 53.4 | 72.8 | |
| SARVisual Pre-training=false, Language Pre-training=false2022.06 | 62.5 | 54.3 | 93.8 | 31.4 | 67.3 | |
| SRNVisual Pre-training=false, Language Pre-training=false2022.06 | 60.1 | 52.3 | 96.7 | 18 | 65 | |
| ASTERVisual Pre-training=false, Language Pre-training=false2022.06 | 54.5 | 52.3 | 93.1 | 38.9 | 64.7 | |
| CRNNVisual Pre-training=false, Language Pre-training=false2022.06 | 53.4 | 54.5 | 97.5 | 46.4 | 67 |