Instance Segmentation On Separated Coco
Metrics
Mean Recall
Results
Performance results of various models on this benchmark
Model Name | Mean Recall | Paper Title | Repository |
|---|---|---|---|
| Swin-B + Cascade Mask R-CNN | 36.31 | Swin Transformer: Hierarchical Vision Transformer using Shifted Windows | |
| Swin-S + Mask R-CNN | 33.67 | Swin Transformer: Hierarchical Vision Transformer using Shifted Windows | |
| Swin-T + Mask R-CNN | 31.94 | Swin Transformer: Hierarchical Vision Transformer using Shifted Windows | |
| Swin-B + Cascade Mask R-CNN (tri-layer modelling) | 36.88 | A Tri-Layer Plugin to Improve Occluded Detection | |
| Swin-S + Mask R-CNN (tri-layer plugin) | 35.80 | A Tri-Layer Plugin to Improve Occluded Detection | |
| Swin-T + Mask R-CNN (tri-layer plugin) | 34.72 | A Tri-Layer Plugin to Improve Occluded Detection |
0 of 6 row(s) selected.