Research map / Detection and segmentation / Open-set and few-shot detection
CRCNet: Few-Shot Segmentation with Cross-Reference and Region–Global Conditional Networks
Weide Liu, Chi Zhang, Guosheng Lin, Fayao Liu
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 cross-reference module to conditional module (ablation) | CRCNet · PASCAL VOC 1-shot | mIoU: 45.8 -> 49.6 | Improved |
| add multi-scale input (ablation) | CRCNet · PASCAL VOC 1-shot | mIoU: 49.6 -> 50.4 | Improved |
| add mask refine module (ablation) | CRCNet · PASCAL VOC 1-shot | mIoU: 50.4 -> 53.0 | Improved |
| add multi-level features (ablation) | CRCNet · PASCAL VOC 1-shot | mIoU: 53.0 -> 56.4 | Improved |
| add global conditional module (ablation) | CRCNet · PASCAL VOC 1-shot | mIoU: 52.0 -> 54.7 | Improved |
| add local conditional module (ablation) | CRCNet · PASCAL VOC 1-shot | mIoU: 52.0 -> 55.3 | Improved |
| combine global and local conditional modules (ablation) | CRCNet · PASCAL VOC 1-shot | mIoU: 52.0 -> 56.4 | Improved |
| finetuning instead of fusion for k-shot (5-shot) (ablation) | CRCNet · PASCAL VOC 5-shot | mIoU: 50.3 -> 54.5 | Improved |
| finetuning instead of fusion (10-shot) (ablation) | CRCNet · PASCAL VOC 10-shot | mIoU: 49.9 -> 58.1 | Improved |
| add fusion on top of finetuning (10-shot) (ablation) | CRCNet · PASCAL VOC 10-shot | mIoU: 58.1 -> 57.8 | Hurt |
| apply finetuning to PFENet (ablation) | PFENet · PASCAL VOC 5-shot | mIoU: 61.9 -> 62.2 | No clear effect |
| sigmoid instead of ReLU activation in cross-reference (ablation) | CRCNet · PASCAL VOC 1-shot | mIoU: 54.6 -> 56.4 | 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.
- CANet — CANet: Class-Agnostic Segmentation Networks With Iterative Refinement and Attentive Few-Shot Learning (CVPR 2019) · better on 3 of 3 shared results
- CRNet* — CRNet: Cross-Reference Networks for Few-Shot Segmentation (CVPR 2020) · 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
- CRNet: Cross-Reference Networks for Few-Shot Segmentation (CVPR 2020 · similarity 0.96)
- Cross-Domain Few-Shot Semantic Segmentation (ECCV 2022 · similarity 0.90)
- A Surprisingly Simple Approach to Generalized Few-Shot Semantic Segmentation (NeurIPS 2024 · similarity 0.87)
- Unlocking the Potential of Pre-trained Vision Transformers for Few-Shot Semantic Segmentation through Relationship Descriptors (CVPR 2024 · similarity 0.87)
- Adapt Before Comparison: A New Perspective on Cross-Domain Few-Shot Segmentation (CVPR 2024 · similarity 0.87)
- The Devil is in Low-Level Features for Cross-Domain Few-Shot Segmentation (CVPR 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.
