Research map / Detection and segmentation / Open-set and few-shot detection
Cross-Domain Few-Shot Semantic Segmentation
Shuo Lei, Xuchao Zhang, Jianfeng He, Fanglan Chen, Bowen Du, Chang-Tien Lu
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 pyramid anchor layers (PATM) for feature transformation (ablation) | PATNet (Res-50) · CD-FSS average, 1-shot | mIoU (%): 47.57 -> 56.06 | Improved |
| add pyramid anchor layers (PATM) for feature transformation (ablation) | PATNet (Res-50) · CD-FSS average, 5-shot | mIoU (%): 51.38 -> 61.99 | Improved |
| share one anchor layer per feature dimension instead of one per feature map (ablation) | PATNet (Res-50) · CD-FSS average, 1-shot | mIoU (%): 54.16 -> 56.06 | Improved |
| share one anchor layer per feature dimension instead of one per feature map (ablation) | PATNet (Res-50) · CD-FSS average, 5-shot | mIoU (%): 59.38 -> 61.99 | Improved |
| fine-tune low-level anchor layer at test time (TFI) (ablation) | PATNet (Res-50) · Deepglobe, 1-shot | mIoU (%): 35.10 -> 37.89 | Improved |
| fine-tune low-level anchor layer at test time (TFI) (ablation) | PATNet (Res-50) · Deepglobe, 5-shot | mIoU (%): 40.72 -> 42.97 | Improved |
| fine-tune high-level anchor layer at test time (ablation) | PATNet (Res-50) · Deepglobe, 1-shot | mIoU (%): 35.10 -> 37.52 | Improved |
| fine-tune medium-level anchor layer at test time (ablation) | PATNet (Res-50) · Deepglobe, 1-shot | mIoU (%): 35.10 -> 34.56 | Hurt |
| fine-tune medium-level anchor layer at test time (ablation) | PATNet (Res-50) · Deepglobe, 5-shot | mIoU (%): 40.72 -> 39.74 | Hurt |
| use PATNet with Res-50 instead of HSNet | few-shot segmentation · CD-FSS average, 1-shot | mIoU (%): 47.57 -> 56.06 | Improved |
| use PATNet instead of HSNet on Chest X-ray | few-shot segmentation · Chest X-ray, 1-shot | mIoU (%): 51.88 -> 66.61 | Improved |
| meta-learned few-shot methods versus linear classifier on frozen Deeplab under large domain gap | HSNet vs LinearDeeplab · Deepglobe, 5-shot | mIoU (%): 35.08 vs 39.69 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PANet* — PANet: Few-Shot Image Semantic Segmentation With Prototype Alignment (ICCV 2019) · 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
- Adapt Before Comparison: A New Perspective on Cross-Domain Few-Shot Segmentation (CVPR 2024 · similarity 0.93)
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (AAAI 2024 · similarity 0.92)
- Unlocking the Potential of Pre-trained Vision Transformers for Few-Shot Semantic Segmentation through Relationship Descriptors (CVPR 2024 · similarity 0.92)
- APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic Segmentation (CVPR 2024 · similarity 0.92)
- The Devil is in Low-Level Features for Cross-Domain Few-Shot Segmentation (CVPR 2025 · similarity 0.91)
- A Surprisingly Simple Approach to Generalized Few-Shot Semantic Segmentation (NeurIPS 2024 · similarity 0.91)
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.
