CateNorm: Categorical Normalization for Robust Medical Image Segmentation
Abstract
Batch normalization (BN) uniformly shifts and scales the activations based on the statistics of a batch of images. However, the intensity distribution of the background pixels often dominates the BN statistics because the background accounts for a large proportion of the entire image. This paper focuses on enhancing BN with the intensity distribution of foreground pixels, the one that really matters for image segmentation. We propose a new normalization strategy, named categorical normalization (CateNorm), to normalize the activations according to categorical statistics. The categorical statistics are obtained by dynamically modulating specific regions in an image that belong to the foreground. CateNorm demonstrates both precise and robust segmentation results across five public datasets obtained from different domains, covering complex and variable data distributions. It is attributable to the ability of CateNorm to capture domain-invariant information from multiple domains (institutions) of medical data. Code is available at https://github.com/lambert-x/CateNorm.
Cite
@article{arxiv.2103.15858,
title = {CateNorm: Categorical Normalization for Robust Medical Image Segmentation},
author = {Junfei Xiao and Lequan Yu and Zongwei Zhou and Yutong Bai and Lei Xing and Alan Yuille and Yuyin Zhou},
journal= {arXiv preprint arXiv:2103.15858},
year = {2022}
}
Comments
Accepted by MICCAI 2022 Workshop on Domain Adaptation and Representation Transfer (DART)