| object-detection-on-coco-o | PVTv2-B5
(Mask R-CNN) | #23 | Average mAP: 28.2Effective Robustness: 6.85 |
| instance-segmentation-on-coco-minival | PVTv2-B4 + Mask R-CNN | #79 | mask AP: 42.7 |
| semantic-segmentation-on-ade20k | PVTv2-B5 + Semantic FPN | #159 | Validation mIoU: 48.7 |
| object-detection-on-coco-val2017 | Sparse R-CNN (PVTv2-B2) | #215 | box AP: 50.1AP50: 69.5AP75: 54.9 |
| object-detection-on-coco-val2017 | Cascade Mask R-CNN (ResNet-50, 3x; reported by PVTv2) | #291 | box AP: 46.3AP50: 64.3AP75: 50.5 |
| object-detection-on-coco-val2017 | GFL (ResNet-50, 3x; reported by PVTv2) | #328 | box AP: 44.5AP50: 63.0AP75: 48.3 |
| object-detection-on-coco-val2017 | ATSS (ResNet-50, 3x; reported by PVTv2) | #341 | box AP: 43.5AP50: 61.9AP75: 47.0 |
| image-classification-on-imagenet | PVTv2-B4 | #416 | Top 1 Accuracy: 83.8%Number of params: 82MGFLOPs: 11.8 |
| image-classification-on-imagenet | PVTv2-B3 | #483 | Top 1 Accuracy: 83.2%Number of params: 45.2MGFLOPs: 6.9 |
| image-classification-on-imagenet | PVTv2-B2 | #619 | Top 1 Accuracy: 82%Number of params: 25.4MGFLOPs: 4 |
| image-classification-on-imagenet | PVTv2-B1 | #848 | Top 1 Accuracy: 78.7%Number of params: 13.1MGFLOPs: 2.1 |
| image-classification-on-imagenet | PVTv2-B0 | #1073 | Top 1 Accuracy: 70.5%Number of params: 3.4MGFLOPs: 0.6 |