Research map / Detection and segmentation / Open-set and few-shot detection
Self-Guided and Cross-Guided Learning for Few-Shot Segmentation
Bingfeng Zhang, Jimin Xiao, Terry Qin
Check 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 SCL to PFENet | PFENet · PASCAL-5^i, 1-shot | mIoU: 60.8 -> 61.8 | Improved |
| Add SCL to PFENet | PFENet · PASCAL-5^i, 5-shot | mIoU: 61.9 -> 62.9 | Improved |
| Add SCL to PFENet | PFENet · COCO-20^i, 1-shot | mIoU: 32.4 -> 37.0 | Improved |
| Add SCL to PFENet | PFENet · COCO-20^i, 5-shot | mIoU: 37.4 -> 39.9 | Improved |
| Add SGM (ablation) | CANet · PASCAL-5^i, 1-shot | mIoU: 55.4 -> 57.5 | Improved |
| Add cross-guided fusion instead of average fusion (ablation) | CANet with SGM · PASCAL-5^i, 5-shot | mIoU: 58.7 -> 59.2 | Improved |
| Use initial vector v_s only instead of the SGM primary and auxiliary vectors (ablation) | SGM · PASCAL-5^i, 1-shot | mIoU: 57.5 -> 55.6 | Hurt |
| Use v_pri only instead of v_pri and v_aux together (ablation) | SGM · PASCAL-5^i, 1-shot | mIoU: 57.5 -> 56.6 | Hurt |
| Use v_aux only instead of v_pri and v_aux together (ablation) | SGM · PASCAL-5^i, 1-shot | mIoU: 57.5 -> 51.4 | Hurt |
| Add initial vector v_s to v_pri and v_aux (ablation) | SGM · PASCAL-5^i, 1-shot | mIoU: 57.5 -> 57.1 | Hurt |
| Remove L_ce^s2 (ablation) | SGM · PASCAL-5^i, 1-shot | mIoU: 57.5 -> 55.6 | Hurt |
| Remove L_ce^s1 (ablation) | SGM · PASCAL-5^i, 1-shot | mIoU: 57.5 -> 56.8 | Hurt |
Closest papers by meaning
- HM: Hybrid Masking for Few-Shot Segmentation (ECCV 2022 · similarity 0.89)
- Few-Shot Segmentation Without Meta-Learning: A Good Transductive Inference Is All You Need? (CVPR 2021 · similarity 0.89)
- Learning What Not To Segment: A New Perspective on Few-Shot Segmentation (CVPR 2022 · similarity 0.89)
- MaskDiff: Modeling Mask Distribution with Diffusion Probabilistic Model for Few-Shot Instance Segmentation (AAAI 2024 · similarity 0.87)
- A Surprisingly Simple Approach to Generalized Few-Shot Semantic Segmentation (NeurIPS 2024 · similarity 0.87)
- Foreground-Covering Prototype Generation and Matching for SAM-Aided Few-Shot Segmentation (AAAI 2025 · 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.
