Frequency Prior Guided Matching: A Data Augmentation Approach for Generalizable Semi-Supervised Polyp Segmentation

TL;DR

This paper introduces Frequency Prior Guided Matching (FPGM), a novel data augmentation method that leverages consistent frequency signatures of polyp edges to improve semi-supervised segmentation models' generalization across diverse datasets.

Contribution

FPGM is the first augmentation framework to utilize frequency priors from polyp edges, enhancing domain invariance and robustness in semi-supervised polyp segmentation.

Findings

Achieves state-of-the-art performance on six public datasets.

Over 10% absolute gain in Dice score in zero-shot generalization.

Significantly improves robustness under limited supervision.

Abstract

Automated polyp segmentation is essential for early diagnosis of colorectal cancer, yet developing robust models remains challenging due to limited annotated data and significant performance degradation under domain shift. Although semi-supervised learning (SSL) reduces annotation requirements, existing methods rely on generic augmentations that ignore polyp-specific structural properties, resulting in poor generalization to new imaging centers and devices. To address this, we introduce Frequency Prior Guided Matching (FPGM), a novel augmentation framework built on a key discovery: polyp edges exhibit a remarkably consistent frequency signature across diverse datasets. FPGM leverages this intrinsic regularity in a two-stage process. It first learns a domain-invariant frequency prior from the edge regions of labeled polyps. Then, it performs principled spectral perturbations on unlabeled images, aligning their amplitude spectra with this learned prior while preserving phase information to maintain structural integrity. This targeted alignment normalizes domain-specific textural variations, thereby compelling the model to learn the underlying, generalizable anatomical structure. Validated on six public datasets, FPGM establishes a new state-of-the-art against ten competing methods. It demonstrates exceptional zero-shot generalization capabilities, achieving over 10% absolute gain in Dice score in data-scarce scenarios. By significantly enhancing cross-domain robustness, FPGM presents a powerful solution for clinically deployable polyp segmentation under limited supervision.