Image Level Supervised Instance Segmentation 2
评估指标
AP
AP@50
AP@75
评测结果
各个模型在此基准测试上的表现结果
| Paper Title | Repository | ||||
|---|---|---|---|---|---|
| WeakSAM-Mask2Former (with SAM) | 25.2 | 38.4 | 27.0 | WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition | |
| WeakSAM-Mask RCNN (with SAM) | 20.6 | 33.9 | 22.0 | WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition | |
| CIM + Mask R-CNN | 17.0 | 29.4 | 17.0 | Complete Instances Mining for Weakly Supervised Instance Segmentation | |
| BESTIE (proposal-free) | 14.3 | 28.0 | 13.2 | Beyond Semantic to Instance Segmentation: Weakly-Supervised Instance Segmentation via Semantic Knowledge Transfer and Self-Refinement | |
| CIM | 11.9 | 22.8 | 11.1 | Complete Instances Mining for Weakly Supervised Instance Segmentation | |
| WS-JDS | 6.1 | 11.7 | 5.5 | Cyclic Guidance for Weakly Supervised Joint Detection and Segmentation | - |
0 of 6 row(s) selected.