WaveRNet: Wavelet-Guided Frequency Learning for Multi-Source Domain-Generalized Retinal Vessel Segmentation

TL;DR

WaveRNet introduces a wavelet-guided frequency learning framework that enhances multi-source domain-generalized retinal vessel segmentation by effectively separating domain-invariant features and refining vessel details.

Contribution

The paper proposes WaveRNet, a novel framework combining wavelet decomposition, domain modulation, and hierarchical refinement to improve retinal vessel segmentation across diverse domains.

Findings

Achieves state-of-the-art generalization on four retinal datasets.

Effectively separates illumination-robust low-frequency and vessel boundary high-frequency features.

Demonstrates robustness against domain shifts caused by illumination and contrast variations.

Abstract

Domain-generalized retinal vessel segmentation is critical for automated ophthalmic diagnosis, yet faces significant challenges from domain shift induced by non-uniform illumination and varying contrast, compounded by the difficulty of preserving fine vessel structures. While the Segment Anything Model (SAM) exhibits remarkable zero-shot capabilities, existing SAM-based methods rely on simple adapter fine-tuning while overlooking frequency-domain information that encodes domain-invariant features, resulting in degraded generalization under illumination and contrast variations. Furthermore, SAM's direct upsampling inevitably loses fine vessel details. To address these limitations, we propose WaveRNet, a wavelet-guided frequency learning framework for robust multi-source domain-generalized retinal vessel segmentation. Specifically, we devise a Spectral-guided Domain Modulator (SDM) that integrates wavelet decomposition with learnable domain tokens, enabling the separation of illumination-robust low-frequency structures from high-frequency vessel boundaries while facilitating domain-specific feature generation. Furthermore, we introduce a Frequency-Adaptive Domain Fusion (FADF) module that performs intelligent test-time domain selection through wavelet-based frequency similarity and soft-weighted fusion. Finally, we present a Hierarchical Mask-Prompt Refiner (HMPR) that overcomes SAM's upsampling limitation through coarse-to-fine refinement with long-range dependency modeling. Extensive experiments under the Leave-One-Domain-Out protocol on four public retinal datasets demonstrate that WaveRNet achieves state-of-the-art generalization performance. The source code is available at https://github.com/Chanchan-Wang/WaveRNet.