Research map / Label-efficient and robust learning / Semi-supervised learning
SSR-SAM: Retrieval-Style Segment Anything Model for Semi-Supervised Ultra-High-Resolution Image Segmentation
Shijie Li, Yiming Chen, Zhineng Chen, Kai Hu, Xieping Gao
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 prompt perturbation on top of image+feature perturbation (ablation) | SSR-SAM · Inria Aerial (63 labeled) | mIoU: 83.1 -> 84.7 | Improved |
| add prompt perturbation on top of image+feature perturbation (ablation) | SSR-SAM · Inria Aerial (252 labeled) | mIoU: 86.2 -> 86.7 | Improved |
| prompt perturbation alone instead of image+feature perturbation (ablation) | SSR-SAM · Inria Aerial (126 labeled) | mIoU: 85.1 -> 85.2 | No clear effect |
| add consistency regularization with perturbations over labeled-only training (ablation) | SSR-SAM · Inria Aerial (63 labeled) | mIoU: 79.5 -> 84.7 | Improved |
| sample student prompts from labeled masks, pseudo-masks, or both (ablation) | SSR-SAM · Inria Aerial (126 labeled) | mIoU: 84.9% to 85.1% | No clear effect |
| use UNI encoder instead of SAM encoder (ablation) | SSR-SAM · BCSS (1/8 labeled) | mIoU: 54.2 -> 57.7 | Improved |
| semi-supervised SSR-SAM over labeled-only SAM | SSR-SAM · URUR (16% labeled) | mIoU: 42.1 -> 44.7 | Improved |
| SSR-SAM versus DINO-based UniMatch v2 | SSR-SAM · URUR (2% labeled) | mIoU: 45.1 -> 43.0 | Hurt |
| SSR-SAM (SAM) versus UniMatch v2 | SSR-SAM · Inria Aerial (16% labeled) | mIoU: 86.1 -> 87.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- UniMatch v2 — UniMatch: Universal Matching from Atom to Task for Few-Shot Drug Discovery (ICLR 2025) · better on 4 of 6 shared results
- SemiVL — SemiVL: Semi-Supervised Semantic Segmentation with Vision-Language Guidance (ECCV 2024) · better on 2 of 2 shared results
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
- Segment Anything without Supervision (NeurIPS 2024 · similarity 0.89)
- Uncertainty-aware Fine-tuning of Segmentation Foundation Models (NeurIPS 2024 · similarity 0.88)
- SAM-PARSER: Fine-Tuning SAM Efficiently by Parameter Space Reconstruction (AAAI 2024 · similarity 0.88)
- BLO-SAM: Bi-level Optimization Based Finetuning of the Segment Anything Model for Overfitting-Preventing Semantic Segmentation (ICML 2024 · similarity 0.88)
- Enhancing SAM with Efficient Prompting and Preference Optimization for Semi-supervised Medical Image Segmentation (CVPR 2025 · similarity 0.88)
- Toward Real Ultra Image Segmentation: Leveraging Surrounding Context to Cultivate General Segmentation Model (NeurIPS 2024 · similarity 0.88)
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.
