CVPR Findings2026
Unify the Views: View-Consistent Prototype Learning for Few-Shot Segmentation
Hongli Liu, Yu Wang, Shengjie Zhao
Abstract
Few-shot segmentation (FSS) has gained significant attention for its ability to generalize to novel classes with limited supervision, yet remains challenged by structural misalignment and cross-view inconsistency under large appearance or viewpoint variations. This paper tackles these challenges by introducing VINE (View-Informed NEtwork), a unified framework that jointly models structural consistency and foreground discrimination to refine class-specific prototypes. Specifically, VINE introduces a spatial-view graph on backbone features, where the spatial graph captures local geometric topology and the view graph connects features from different perspectives to propagate view-invariant structural semantics. To further alleviate foreground ambiguity, we derive a discriminative prior from the support-query feature discrepancy to capture category-specific contrast, which reweights SAM features by emphasizing salient regions and recalibrates backbone activations for improved structural focus. The foreground-enhanced SAM features and structurally enriched ResNet features are progressively integrated through masked cross-attention, yielding class-consistent prototypes used as adaptive prompts for the SAM decoder to generate accurate masks. Extensive experiments on multiple FSS benchmarks validate the effectiveness and robustness of VINE, particularly under challenging scenarios with viewpoint shifts and complex structures. The code is available at
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Learning What Not to Segment: A New Perspective on Few-Shot Segmentation cites Pyramid Graph Networks With Connection Attentions for Region-Based One-Shot Semantic Segmentation
- Learning What Not to Segment: A New Perspective on Few-Shot Segmentation cites PANet: Few-Shot Image Semantic Segmentation With Prototype Alignment
- Learning What Not to Segment: A New Perspective on Few-Shot Segmentation cites Feature Weighting and Boosting for Few-Shot Segmentation
Builds on21
- Segment AnythingAlexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao et al.ICCV 2023 · 13,211 citations
- PANet: Few-Shot Image Semantic Segmentation With Prototype AlignmentKaixin Wang, Jun Hao Liew, Yingtian Zou, Daquan Zhou et al.ICCV 2019 · 1,404 citations
- Feature Weighting and Boosting for Few-Shot SegmentationKhoi Nguyen, Sinisa TodorovicICCV 2019 · 402 citations
- Pyramid Graph Networks With Connection Attentions for Region-Based One-Shot Semantic SegmentationChi Zhang, Guosheng Lin, Fayao Liu, Jiushuang Guo et al.ICCV 2019 · 351 citations
- Learning What Not to Segment: A New Perspective on Few-Shot SegmentationChunbo Lang, Gong Cheng, Binfei Tu, Junwei HanCVPR 2022 · 289 citations
Related papers
- Relevant Intrinsic Feature Enhancement Network for Few-Shot Semantic SegmentationXiaoyi Bao, Jie Qin, Siyang Sun, Xingang Wang et al.AAAI 2024 · 34 citations
- Foreground-Covering Prototype Generation and Matching for SAM-Aided Few-Shot SegmentationSuho Park, SuBeen Lee, Hyun Seok Seong, Jaejoon Yoo et al.AAAI 2025 · 9 citations
- Scale-Aware Graph Neural Network for Few-Shot Semantic SegmentationGuo-Sen Xie, Jie Liu, Huan Xiong, Ling ShaoCVPR 2021
- SAM-Aware Graph Prompt Reasoning Network for Cross-Domain Few-Shot SegmentationShi-Feng Peng, Guolei Sun, Yong Li, Hongsong Wang et al.AAAI 2025 · 7 citations
- Towards Complete Activation: Foreground-Background Multi-Perspective Guided Cross-Support for Few-Shot SegmentationYi Yang, Qiang Jiao, Mengrui Shi, Qiang ZhangCVPR Findings 2026
