Unsupervised Semantic Segmentation On Pascal 1
Metrics
Clustering [mIoU]
FCN [mIoU]
Results
Performance results of various models on this benchmark
Model Name | Clustering [mIoU] | FCN [mIoU] | Paper Title | Repository |
|---|---|---|---|---|
| Leopart (ViT-B/8) | 47.2 | 76.3 | Self-Supervised Learning of Object Parts for Semantic Segmentation | |
| SegSort (Edges) | - | - | SegSort: Segmentation by Discriminative Sorting of Segments | |
| CAUSE (ViT-B/8) | 53.3 | - | Causal Unsupervised Semantic Segmentation | |
| Leopart (ViT-S/16) | 41.7 | 71.4 | Self-Supervised Learning of Object Parts for Semantic Segmentation | |
| CAUSE (iBOT, ViT-B/16) | 53.4 | - | Causal Unsupervised Semantic Segmentation | |
| MaskDistill | 45.8 | - | Discovering Object Masks with Transformers for Unsupervised Semantic Segmentation | |
| MaskDistill+CRF | 48.9 | - | Discovering Object Masks with Transformers for Unsupervised Semantic Segmentation | |
| CAUSE (DINOv2, ViT-B/14) | 53.2 | - | Causal Unsupervised Semantic Segmentation | |
| MaskContrast (Saliency) | 44.2 | - | Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals | |
| MaskContrast | 35.0 | - | Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals |
0 of 10 row(s) selected.