Unsupervised Saliency Detection 벤치마크
Unsupervised Saliency Detection on DUTS
maximal F-measure
- 2022-02-23 — TokenCut: maximal F-measure 75.5
- 2022-03-23 — SelfMask: maximal F-measure 88.2
| Rank | Model | maximal F-measure | Accuracy | IoU | Paper | Code | Year |
|---|---|---|---|---|---|---|---|
| 1 | SelfMask | 88.2 | 93.3 | 66 | Unsupervised Salient Object Detection with Spectral Cluster Voting | noelshin/selfmask | 2022 |
| 2 | MOVE | 82.9 | 95.4 | 72.8 | MOVE: Unsupervised Movable Object Segmentation and Detection | adambielski/move-seg | 2022 |
| 3 | TokenCut | 75.5 | 91.4 | 62.4 | Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut | YangtaoWANG95/TokenCut | 2022 |