Research map / Detection and segmentation / Open-set and few-shot detection
Training-free Boosting for Few-shot Segmentation via Generalizing Semantic Mining
Kangyu Xiao, Zilei Wang, Yixin Zhang, Junjie Li
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 GSM (TSR+QSM+RR) to GF-SAM | GF-SAM · COCO-20i 1-shot | mIoU: 58.7 -> 61.2 | Improved |
| add GSM to GF-SAM | GF-SAM · LVIS-92i 1-shot | mIoU: 35.2 -> 38.9 | Improved |
| add GSM to TBS (specialist) | TBS · Pascal-5i 1-shot | mIoU: 71.2 -> 74.4 | Improved |
| add GSM to Matcher | Matcher · ISIC 5-shot (cross-domain) | mIoU: 35.0 -> 44.5 | Improved |
| add GSM to Matcher | Matcher · FSS-1000 5-shot (cross-domain) | mIoU: 89.6 -> 89.6 | No clear effect |
| add Target Semantic Reusing alone (m_pred as proxy mask) (ablation) | GF-SAM · ID-FSS average, 1-shot | mIoU: 55.3 -> 56.5 | Improved |
| add TSR alone (ablation) | GF-SAM · ID-FSS average, 5-shot | mIoU: 64.5 -> 64.2 | Hurt |
| add QSM on top of TSR (ablation) | GF-SAM + TSR · ID-FSS average, 1-shot | mIoU: 56.5 -> 58.6 | Improved |
| add Representative Re-weighting (ablation) | GF-SAM · ID-FSS average, 1-shot | mIoU: 55.3 -> 55.3 | No clear effect |
| add Representative Re-weighting (ablation) | GF-SAM · ID-FSS average, 5-shot | mIoU: 64.5 -> 65.3 | Improved |
| add RR on top of TSR+QSM (ablation) | GF-SAM + TSR + QSM · CD-FSS average, 5-shot | mIoU: 64.8 -> 65.2 | Improved |
| add GSM to GF-SAM | GF-SAM · Inference cost | s/img: 1.88 -> 2.13 | Hurt |
Closest papers by meaning
- A Surprisingly Simple Approach to Generalized Few-Shot Semantic Segmentation (NeurIPS 2024 · similarity 0.89)
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (AAAI 2024 · similarity 0.88)
- Unlocking the Potential of Pre-trained Vision Transformers for Few-Shot Semantic Segmentation through Relationship Descriptors (CVPR 2024 · similarity 0.88)
- Enhancing Generalized Few-Shot Semantic Segmentation via Effective Knowledge Transfer (AAAI 2025 · similarity 0.88)
- DSV-LFS: Unifying LLM-Driven Semantic Cues with Visual Features for Robust Few-Shot Segmentation (CVPR 2025 · similarity 0.88)
- APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic 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.
