Research map / Detection and segmentation / Open-set and few-shot detection
Dynamic Prototype Convolution Network for Few-Shot Semantic 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 |
|---|---|---|---|
| add DPCN's SAM, FFM, and DCM to the baseline | FSS with ResNet50 · PASCAL-5^i 1-shot | mIoU: 61.4 -> 66.7 | Helped |
| add DPCN's SAM, FFM, and DCM to the baseline | FSS with ResNet50 · COCO-20^i 1-shot | mIoU: 34.5 -> 43.0 | Helped |
| add DCM (ablation) | DPCN with SAM and FFM · PASCAL-5^i 1-shot | mIoU: 64.5 -> 66.7 | Helped |
| add FFM (ablation) | DPCN with SAM and DCM · PASCAL-5^i 1-shot | mIoU: 65.4 -> 66.7 | Helped |
| add SAM (ablation) | DPCN with FFM and DCM · PASCAL-5^i 1-shot | mIoU: 64.7 -> 66.7 | Helped |
| use DCM kernel size 5 instead of 3, 7, or 9 (ablation) | DPCN · PASCAL-5^i 1-shot | mIoU: 65.9; 66.7; 66.1; 66.3 | Mixed |
| combine square and asymmetric DCM kernels (ablation) | DPCN · PASCAL-5^i 1-shot | mIoU: 65.0; 65.2; 64.7 -> 66.7 | Helped |
| add DCM to CANet (ablation) | CANet · PASCAL-5^i 1-shot | mIoU: 55.4 -> 58.8 | Helped |
| add DCM to PFENet (ablation) | PFENet · PASCAL-5^i 1-shot | mIoU: 60.8 -> 62.3 | Helped |
| generate asymmetric and square kernels in parallel rather than serially (ablation) | DPCN DCM · PASCAL-5^i 1-shot | mIoU: 66.7 -> 66.0 | Hurt |
| use multi-scale testing (ablation) | DPCN with ResNet50 · PASCAL-5^i 1-shot | mIoU: 65.7 -> 66.7 | Helped |
| use multi-scale testing (ablation) | DPCN with ResNet50 · COCO-20^i 1-shot | mIoU: 41.5 -> 43.0 | Helped |
Closest papers by meaning
- Foreground-Covering Prototype Generation and Matching for SAM-Aided Few-Shot Segmentation (AAAI 2025 · similarity 0.90)
- DSV-LFS: Unifying LLM-Driven Semantic Cues with Visual Features for Robust Few-Shot Segmentation (CVPR 2025 · similarity 0.89)
- Adaptive FSS: A Novel Few-Shot Segmentation Framework via Prototype Enhancement (AAAI 2024 · similarity 0.88)
- APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic Segmentation (CVPR 2024 · similarity 0.87)
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (AAAI 2024 · similarity 0.87)
- DyPolySeg: Taylor Series-Inspired Dynamic Polynomial Fitting Network for Few-shot Point Cloud Semantic Segmentation (ICML 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.