Boosting SAM for Cross-Domain Few-Shot Segmentation via Conditional Point Sparsification

2026-02-05

AI Summary

This work addresses the significant performance degradation of Segment Anything Model (SAM) in cross-domain few-shot segmentation due to domain shift, which adversely affects its dense matching points. To mitigate this issue, the authors propose Conditional Point Sparsification (CPS), a training-free approach that adaptively sparsifies matching points based on the ground-truth mask of the reference image, thereby guiding SAM to produce more accurate segmentation results. This study is the first to reveal the critical influence of point density on cross-domain few-shot segmentation and introduces an adaptive, training-free point sparsification mechanism that effectively alleviates the interference of domain shift on SAM’s prompt-based interaction. Extensive experiments demonstrate that the proposed method substantially outperforms existing training-free SAM approaches across multiple cross-domain datasets, achieving notably higher segmentation accuracy.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Motivated by the success of the Segment Anything Model (SAM) in promptable segmentation, recent studies leverage SAM to develop training-free solutions for few-shot segmentation, which aims to predict object masks in the target image based on a few reference exemplars. These SAM-based methods typically rely on point matching between reference and target images and use the matched dense points as prompts for mask prediction. However, we observe that dense points perform poorly in Cross-Domain Few-Shot Segmentation (CD-FSS), where target images are from medical or satellite domains. We attribute this issue to large domain shifts that disrupt the point-image interactions learned by SAM, and find that point density plays a crucial role under such conditions. To address this challenge, we propose Conditional Point Sparsification (CPS), a training-free approach that adaptively guides SAM interactions for cross-domain images based on reference exemplars. Leveraging ground-truth masks, the reference images provide reliable guidance for adaptively sparsifying dense matched points, enabling more accurate segmentation results. Extensive experiments demonstrate that CPS outperforms existing training-free SAM-based methods across diverse CD-FSS datasets.

Problem

• Cross-Domain Few-Shot Segmentation
• Segment Anything Model
• domain shift
• point prompting
• few-shot segmentation

Innovation

• Conditional Point Sparsification
• Cross-Domain Few-Shot Segmentation
• Segment Anything Model
• Training-Free Adaptation
• Point Sparsification