| DynaSeg: A Deep Dynamic Fusion Method for Unsupervised Image Segmentation Incorporating Feature Similarity and Spatial Continuity | ✓ Link | 54.1 | 81.1 | | | DynaSeg - FSF (ResNet-18 FPN) | 2024-05-09 |
| DiffCut: Catalyzing Zero-Shot Semantic Segmentation with Diffusion Features and Recursive Normalized Cut | ✓ Link | 49.1 | | | | DiffCut | 2024-06-05 |
| Causal Unsupervised Semantic Segmentation | ✓ Link | 45.3 | 78.0 | | | CAUSE (DINOv2, ViT-B/14) | 2023-10-11 |
| Causal Unsupervised Semantic Segmentation | ✓ Link | 41.9 | 74.9 | | | CAUSE (ViT-B/8) | 2023-10-11 |
| Unsupervised Universal Image Segmentation | ✓ Link | 30.2 | 63.9 | | | U2Seg | 2023-12-28 |
| Boosting Unsupervised Semantic Segmentation with Principal Mask Proposals | ✓ Link | 29.7 | 57.9 | | | PriMaPs+STEGO (DINO ViT-B/8) | 2024-04-25 |
| Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling | ✓ Link | 29.0 | 58.6 | 41.6 | 75.5 | DepthG (ViT-B/8) | 2023-09-21 |
| Unsupervised Semantic Segmentation by Distilling Feature Correspondences | ✓ Link | 28.2 | 56.9 | | | STEGO (ViT-B/8) | 2022-03-16 |
| EAGLE: Eigen Aggregation Learning for Object-Centric Unsupervised Semantic Segmentation | ✓ Link | 27.2 | 64.2 | 43.9 | 76.8 | EAGLE (DINO, ViT-S/8) | 2024-03-03 |
| []() | | 26.7 | 63.2 | | | SmooSeg (DINO, ViT-S/8) | |
| Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling | ✓ Link | 26.7 | 55.1 | 37.8 | 73.9 | DepthG w/ 3D-LHP (ViT-S/8) | 2023-09-21 |
| Expand-and-Quantize: Unsupervised Semantic Segmentation Using High-Dimensional Space and Product Quantization | | 25.8 | 53.8 | 41.2 | 75.2 | EQUSS | 2023-12-12 |
| Expand-and-Quantize: Unsupervised Semantic Segmentation Using High-Dimensional Space and Product Quantization | | 25.8 | 53.8 | | | EQUSS (ViT-S) | 2023-12-12 |
| Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling | ✓ Link | 25.6 | 56.3 | 38.9 | 73.7 | DepthG (ViT-S/8) | 2023-09-21 |
| Boosting Unsupervised Semantic Segmentation with Principal Mask Proposals | ✓ Link | 25.1 | 57.8 | | | PriMaPs+HP (DINO ViT-S/8) | 2024-04-25 |
| Leveraging Hidden Positives for Unsupervised Semantic Segmentation | ✓ Link | 24.6 | 57.2 | 42.7 | 75.6 | HP (ViT-S/8) | 2023-03-27 |
| Unsupervised Semantic Segmentation by Distilling Feature Correspondences | ✓ Link | 24.5 | | 38.3 | 74.4 | STEGO (ViT-S/8) | 2022-03-16 |
| Leveraging Hidden Positives for Unsupervised Semantic Segmentation | ✓ Link | 24.3 | 54.5 | 39.1 | 74.1 | HP (ViT-S/16) | 2023-03-27 |
| CrOC: Cross-View Online Clustering for Dense Visual Representation Learning | ✓ Link | 21.9 | | | | CrOC (ViT-S/16, COCO+) | 2023-03-23 |
| ViCE: Improving Dense Representation Learning by Superpixelization and Contrasting Cluster Assignment | ✓ Link | 21.77 | 64.8 | | | ViCE | 2021-11-24 |
| Fully Self-Supervised Learning for Semantic Segmentation | | 15.69 | 40.38 | | | FS4 | 2022-02-24 |
| PiCIE: Unsupervised Semantic Segmentation using Invariance and Equivariance in Clustering | ✓ Link | 14.36 | 49.99 | | | PiCIE + H | 2021-03-30 |
| Unsupervised Image Semantic Segmentation through Superpixels and Graph Neural Networks | | | 55.7 | | | SGSeg | 2022-10-21 |
| Rethinking Alignment and Uniformity in Unsupervised Semantic Segmentation | | | 52.0 | | | SAN | 2022-11-26 |
| PiCIE: Unsupervised Semantic Segmentation using Invariance and Equivariance in Clustering | ✓ Link | | 48.1 | | | PiCIE | 2021-03-30 |
| Self-Supervised Visual Representation Learning with Semantic Grouping | ✓ Link | | 42.36 | | | Ours (SlotCon) | 2022-05-30 |
| Invariant Information Clustering for Unsupervised Image Classification and Segmentation | ✓ Link | | 21.8 | | | IIC | 2018-07-17 |
| Hierarchical Context Learning of object components for unsupervised semantic segmentation | ✓ Link | | | 47.4 | 77.3 | HCL (ViT-S/8) | 2025-04-29 |
| Hierarchical Context Learning of object components for unsupervised semantic segmentation | ✓ Link | | | 45.8 | 76.7 | HCL (ViT-S/16) | 2025-04-29 |