Research map / Detection and segmentation / Open-set and few-shot detection
Adaptive Prototype Learning and Allocation for Few-Shot Segmentation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use ASGNet instead of PFENet | Few-shot segmentation with ResNet-101 · Pascal-5^i, 5-shot | mIoU: 61.40 -> 64.36 | Helped |
| Use ASGNet instead of PFENet | Few-shot segmentation on COCO-20^i · COCO-20^i, 5-shot | mIoU: 37.40 -> 42.48 | Helped |
| Set the maximum prototype count to 5 rather than 1 (ablation) | ASGNet · Pascal-5^i, 1-shot | mIoU: 58.74 -> 59.29 | Helped |
| Increase the maximum prototype count beyond 5 to 9 (ablation) | ASGNet · Pascal-5^i, 1-shot | mIoU: 59.29 -> 58.88 | Hurt |
| Use adaptive rather than fixed five-centroid counts (ablation) | ASGNet · Pascal-5^i, 1-shot | mIoU: 59.05 -> 59.29 | Helped |
| Replace masked average pooling with SGC (ablation) | Prototype generation with prototype expansion · Pascal-5^i, 1-shot | mIoU: 58.96 -> 58.31 | Hurt |
| Replace masked average pooling with SGC (ablation) | Prototype generation with prototype expansion · Pascal-5^i, 5-shot | mIoU: 60.19 -> 61.24 | Helped |
| Use guided prototype allocation instead of prototype expansion (ablation) | SGC prototype generation · Pascal-5^i, 1-shot | mIoU: 58.31 -> 59.29 | Helped |
| Use guided prototype allocation instead of prototype expansion (ablation) | SGC prototype generation · Pascal-5^i, 5-shot | mIoU: 61.24 -> 63.94 | Helped |
| Collect prototypes from each support shot rather than attention-weighted fusion (ablation) | ASGNet k-shot strategy · Pascal-5^0, 5-shot | mIoU: 60.69 -> 63.66 | Helped |
| Collect prototypes from each support shot rather than feature averaging (ablation) | ASGNet k-shot strategy · Pascal-5^0, 5-shot | mIoU: 62.10 -> 63.66 | Helped |
Closest papers by meaning
- Learning from the Target: Dual Prototype Network for Few Shot Semantic Segmentation (AAAI 2022 · similarity 0.87)
- Learning What Not To Segment: A New Perspective on Few-Shot Segmentation (CVPR 2022 · similarity 0.86)
- Foreground-Covering Prototype Generation and Matching for SAM-Aided Few-Shot Segmentation (AAAI 2025 · similarity 0.85)
- Adaptive FSS: A Novel Few-Shot Segmentation Framework via Prototype Enhancement (AAAI 2024 · similarity 0.85)
- Bridge the Points: Graph-based Few-shot Segment Anything Semantically (NeurIPS 2024 · similarity 0.85)
- A Surprisingly Simple Approach to Generalized Few-Shot Semantic Segmentation (NeurIPS 2024 · similarity 0.84)
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.