XNet v2: Fewer Limitations, Better Results and Greater Universality

TL;DR

XNet v2 introduces a wavelet-based, multi-frequency fusion architecture that significantly improves semi-supervised biomedical segmentation, overcoming limitations of the original XNet and achieving state-of-the-art results across multiple datasets.

Contribution

The paper presents XNet v2, a novel wavelet-based fusion model that enhances low- and high-frequency information transfer, improving segmentation performance and universality.

Findings

Achieves state-of-the-art semi-supervised segmentation results.

Demonstrates superior performance on multiple 2D and 3D datasets.

Outperforms original XNet in scenarios with limited high-frequency information.

Abstract

XNet introduces a wavelet-based X-shaped unified architecture for fully- and semi-supervised biomedical segmentation. So far, however, XNet still faces the limitations, including performance degradation when images lack high-frequency (HF) information, underutilization of raw images and insufficient fusion. To address these issues, we propose XNet v2, a low- and high-frequency complementary model. XNet v2 performs wavelet-based image-level complementary fusion, using fusion results along with raw images inputs three different sub-networks to construct consistency loss. Furthermore, we introduce a feature-level fusion module to enhance the transfer of low-frequency (LF) information and HF information. XNet v2 achieves state-of-the-art in semi-supervised segmentation while maintaining competitve results in fully-supervised learning. More importantly, XNet v2 excels in scenarios where XNet fails. Compared to XNet, XNet v2 exhibits fewer limitations, better results and greater universality. Extensive experiments on three 2D and two 3D datasets demonstrate the effectiveness of XNet v2. Code is available at https://github.com/Yanfeng-Zhou/XNetv2 .