Unsupervised Semantic Segmentation
Models that learn to segment each image (i.e. assign a class to every pixel) without seeing the ground truth labels.
( Image credit: SegSort: Segmentation by Discriminative Sorting of Segments )
Papers
Showing 1–10 of 95 papers
All datasetsCOCO-Stuff-27Cityscapes testPASCAL VOC 2012 valPotsdam-3COCO-Stuff-3ImageNet-S-50COCO-Stuff-171COCO-Stuff-81SUIMCOCO-Stuff-15ACDC (Adverse Conditions Dataset with Correspondences)Cityscapes val
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | GraPix | Pixel Accuracy | 64.89 | — | Unverified |
| 2 | CUPS | mIoU | 26.8 | — | Unverified |
| 3 | ViCE | mIoU | 25.2 | — | Unverified |
| 4 | EAGLE (DINO, ViT-B/8) | mIoU | 22.1 | — | Unverified |
| 5 | EQUSS | mIoU | 22 | — | Unverified |
| 6 | PriMaPs-EM + STEGO (DINO ViT-B/8) | mIoU | 21.6 | — | Unverified |
| 7 | STEGO | mIoU | 21 | — | Unverified |
| 8 | EAGLE (DINO, ViT-S/8) | mIoU | 19.7 | — | Unverified |
| 9 | PriMaPs-EM (DINO ViT-S/8) | mIoU | 19.4 | — | Unverified |
| 10 | HP | mIoU | 18.4 | — | Unverified |