Divide-and-Conquer Decoupled Network for Cross-Domain Few-Shot Segmentation

TL;DR

This paper introduces DCDNet, a novel network for cross-domain few-shot segmentation that decouples features into category and domain components, improving generalization and adaptation across unseen domains.

Contribution

The paper proposes a divide-and-conquer decoupled network with novel modules for feature disentanglement and adaptive fusion, advancing cross-domain few-shot segmentation performance.

Findings

Outperforms existing methods on four datasets.

Achieves state-of-the-art cross-domain generalization.

Enhances few-shot adaptation capabilities.

Abstract

Cross-domain few-shot segmentation (CD-FSS) aims to tackle the dual challenge of recognizing novel classes and adapting to unseen domains with limited annotations. However, encoder features often entangle domain-relevant and category-relevant information, limiting both generalization and rapid adaptation to new domains. To address this issue, we propose a Divide-and-Conquer Decoupled Network (DCDNet). In the training stage, to tackle feature entanglement that impedes cross-domain generalization and rapid adaptation, we propose the Adversarial-Contrastive Feature Decomposition (ACFD) module. It decouples backbone features into category-relevant private and domain-relevant shared representations via contrastive learning and adversarial learning. Then, to mitigate the potential degradation caused by the disentanglement, the Matrix-Guided Dynamic Fusion (MGDF) module adaptively integrates base, shared, and private features under spatial guidance, maintaining structural coherence. In addition, in the fine-tuning stage, to enhanced model generalization, the Cross-Adaptive Modulation (CAM) module is placed before the MGDF, where shared features guide private features via modulation ensuring effective integration of domain-relevant information. Extensive experiments on four challenging datasets show that DCDNet outperforms existing CD-FSS methods, setting a new state-of-the-art for cross-domain generalization and few-shot adaptation.