Research map / Detection and segmentation / Open-set and few-shot detection
Learning Non-Target Knowledge for Few-Shot Semantic Segmentation
Yuanwei Liu, Nian Liu, Qinglong Cao, Xiwen Yao, Junwei Han, Ling Shao
This paper tried a common change: Add a contrastive loss
Add prototypical contrastive learning to BGEM + DOEM · PASCAL-5^i 1-shot · mIoU: 63.6 -> 64.2 Improved
What 1,317 accepted papers report for this change
- 1,253 improved
- 22 no clear effect
- 16 mixed
- 26 hurt
- Beyond Looking Up, Try Looking Around: Harmonizing Global Structure and Local Consistency in Optimal Transport for Short Text Clustering ICML 2026 · add contrastive learning · SearchSnippets · ACC: 56.65 -> 80.12Improved
- SleepLM: Natural-Language Intelligence for Human Sleep ICML 2026 · add contrastive objective to captioning (CoCa vs Cap) · sleep event IoU · IoU: 29.2 -> 27.4Hurt
- LIMSSR: LLM-Driven Sequence-to-Score Reasoning under Training-Time Incomplete Multimodal Observations ICML 2026 · add contrastive loss L_con · FS1000 average · Spearman correlation: 0.759 -> 0.789Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 |
|---|---|---|---|
| Eliminate BG regions with BGEM (ablation) | NTRENet · PASCAL-5^i 1-shot | mIoU: 60.0 -> 62.4 | Improved |
| Eliminate DO regions with DOEM after BGEM (ablation) | NTRENet · PASCAL-5^i 1-shot | mIoU: 62.4 -> 63.6 | Improved |
| Add prototypical contrastive learning to BGEM + DOEM (ablation) | NTRENet · PASCAL-5^i 1-shot | mIoU: 63.6 -> 64.2 | Improved |
| Eliminate BG regions with BGEM (ablation) | NTRENet · PASCAL-5^i 1-shot | precision: 61.9 -> 62.8 | Improved |
| Eliminate DO regions with DOEM after BGEM (ablation) | NTRENet · PASCAL-5^i 1-shot | precision: 62.8 -> 63.3 | Improved |
| Add prototypical contrastive learning to BGEM + DOEM (ablation) | NTRENet · PASCAL-5^i 1-shot | precision: 63.3 -> 63.6 | Improved |
| Use DO prototypes rather than whole non-target prototypes as PCL negatives (ablation) | NTRENet · PASCAL-5^i 1-shot | mIoU: 61.8 -> 64.2 | Improved |
| Use DO prototypes rather than BG prototypes as PCL negatives (ablation) | NTRENet · PASCAL-5^i 1-shot | mIoU: 62.6 -> 64.2 | Improved |
| Vary BG prototype channel dimension (ablation) | NTRENet · PASCAL-5^i 1-shot | mIoU | Mixed |
| Mine and eliminate non-target regions rather than directly focusing only on target segmentation | NTRENet · PASCAL-5^i 1-shot | mIoU: 61.8 -> 64.2 | Improved |
| Mine and eliminate non-target regions rather than directly focusing only on target segmentation | NTRENet · COCO-20^i 1-shot | mIoU: 34.6 -> 39.3 | Improved |
Closest papers by meaning
- Bayesian Decomposition and Semantic Completion for Few-shot Semantic Segmentation (CVPR 2026 · similarity 0.89)
- Generalized Few-Shot Semantic Segmentation (CVPR 2022 · similarity 0.89)
- Cross-Domain Few-Shot Semantic Segmentation (ECCV 2022 · similarity 0.88)
- A Surprisingly Simple Approach to Generalized Few-Shot Semantic Segmentation (NeurIPS 2024 · similarity 0.87)
- Task-Disruptive Background Suppression for Few-Shot Segmentation (AAAI 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)
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.
