Hierarchical Spatial and Channel Aggregation for Cross-domain Few-shot Segmentation

2026-06-23

AI Summary

This work addresses the degradation of support-query matching in cross-domain few-shot segmentation, which arises from discrepancies in semantic granularity and discriminative attributes between domains. To tackle this challenge, the authors propose a Dual Hierarchical Aggregation Network (DHANet) that enhances semantic representation through multi-scale region aggregation in the spatial dimension and improves discriminability via multi-scale attribute aggregation in the channel dimension. Furthermore, they introduce, for the first time, an Online Probabilistic Semantic Bank (OPSB) that dynamically generates pseudo-prototypes during inference to alleviate the scarcity of support samples. By jointly modeling cross-domain semantic and attribute disparities, the proposed method achieves state-of-the-art performance across four target-domain datasets and significantly mitigates the over-alignment issue in prototype matching.

Abstract

Cross-domain Few-shot Segmentation (CD-FSS) aims to learn generalizable segmentation capability from abundant annotated samples in the source domain, enabling accurate segmentation of novel classes in the target domain with only a few annotated samples. Existing CD-FSS methods mainly focus on mitigating feature distribution shifts caused by style gaps while ignoring significant differences in class semantic granularity and discriminative attributes across domains, leading to two key degradations in support-query matching: semantic over-alignment and attribute over-alignment. To this end, we propose the Dual Hierarchical Aggregation Network (DHANet), which comprises three key modules. First, the Hierarchical Spatial Aggregation (HSA) module performs multi-scale region aggregation of pixel features along the spatial dimension, generating hierarchical semantic-enhanced features to alleviate semantic over-alignment. Additionally, the HCA module conducts multi-scale attribute aggregation along the channel dimension, generating hierarchical attribute-enhanced features to mitigate attribute over-alignment. Finally, we propose the Online Probabilistic Semantic Bank (OPSB), which progressively constructs and updates class probability distributions from query predictions during inference, and samples multiple pseudo-prototypes as additional support information to mitigate insufficient support. Extensive experiments on four target-domain datasets demonstrate that our method achieves state-of-the-art performance.

Problem

• Cross-domain Few-shot Segmentation
• Semantic Over-alignment
• Attribute Over-alignment
• Domain Shift
• Few-shot Learning

Innovation

• Hierarchical Spatial Aggregation
• Hierarchical Channel Aggregation
• Cross-domain Few-shot Segmentation
• Semantic Over-alignment
• Online Probabilistic Semantic Bank