Research map / Detection and segmentation / Weakly supervised detection and segmentation
Large-Scale Unsupervised Semantic Segmentation
Shanghua Gao, Zhong-Yu Li, Ming–Hsuan Yang, Ming‐Ming Cheng, Junwei Han, Philip H. S. Torr
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add non-contrastive P2P alignment (ablation) | SwAV · ImageNet-S300 (distance matching) | test mIoU: 22.6 -> 24.8 | Improved |
| add clustering P2P alignment (ablation) | SwAV · ImageNet-S300 | test mIoU: 22.6 -> 21.2 | Hurt |
| add contrastive P2P alignment (ablation) | SwAV · ImageNet-S300 | test mIoU: 22.6 -> 18.0 | Hurt |
| add D2S supervision at stage 3 (ablation) | SwAV+P2P · ImageNet-S300 | test mIoU: 24.8 -> 25.2 | Improved |
| supervise stages 3-2 instead of stage 3 (ablation) | SwAV+P2P · ImageNet-S300 | test mIoU: 25.2 -> 24.9 | Hurt |
| add D2S supervision at stages 3-2 (ablation) | PixelPro+P2P · ImageNet-S300 | test mIoU: 22.0 -> 23.4 | Improved |
| add clustering loss (ablation) | PixelPro · ImageNet-S300 | test mIoU: 15.8 -> 21.3 | Improved |
| use projection M in P2P (ablation) | SwAV+P2P · ImageNet-S300 | Img-Acc: 57.1 -> 58.5 | Improved |
| use deep-to-shallow instead of same-stage supervision (ablation) | PixelPro+P2P · ImageNet-S300 | test mIoU: 22.6 -> 23.4 | Improved |
| cross-view instead of same-view D2S supervision (ablation) | PixelPro+P2P · ImageNet-S300 | test mIoU: 23.1 -> 23.4 | Improved |
| pixel-attention label generation instead of pixel-level (ablation) | PASS_s · ImageNet-S50 | test mIoU: 12.7 -> 29.3 | Improved |
| add fine-tuning step (ablation) | PASS_s · ImageNet-S50 | test mIoU: 26.0 -> 29.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PiCIE — PiCIE: Unsupervised Semantic Segmentation Using Invariance and Equivariance in Clustering (CVPR 2021) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (AAAI 2024 · similarity 0.89)
- Hierarchy-Agnostic Unsupervised Segmentation: Parsing Semantic Image Structure (NeurIPS 2024 · similarity 0.88)
- Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling (CVPR 2024 · similarity 0.88)
- Integrating Low-Level Visual Cues for Enhanced Unsupervised Semantic Segmentation (AAAI 2025 · similarity 0.87)
- Segment Anything without Supervision (NeurIPS 2024 · similarity 0.87)
- Unsupervised Universal Image Segmentation (CVPR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
