Single domain generalization via normalised cross-correlation based convolutions
Abstract
Deep learning techniques often perform poorly in the presence of domain shift, where the test data follows a different distribution than the training data. The most practically desirable approach to address this issue is Single Domain Generalization (S-DG), which aims to train robust models using data from a single source. Prior work on S-DG has primarily focused on using data augmentation techniques to generate diverse training data. In this paper, we explore an alternative approach by investigating the robustness of linear operators, such as convolution and dense layers commonly used in deep learning. We propose a novel operator called "XCNorm"that computes the normalized cross-correlation between weights and an input feature patch. This approach is invariant to both affine shifts and changes in energy within a local feature patch and eliminates the need for commonly used non-linear activation functions. We show that deep neural networks composed of this operator are robust to common semantic distribution shifts. Furthermore, our empirical results on single-domain generalization benchmarks demonstrate that our proposed technique performs comparably to the stateof-the-art methods.
RAS ID
70479
Document Type
Conference Proceeding
Date of Publication
1-3-2024
Funding Information
Australian Government / Ford Motor Company
School
School of Science
Grant Number
LP190100165
Copyright
subscription content
Publisher
IEEE
Comments
Chuah, W., Tennakoon, R., Hoseinnezhad, R., Suter, D., & Bab-Hadiashar, A. (2024). Single domain generalization via normalised cross-correlation based convolutions. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (pp. 1741-1750). https://doi.org/10.1109/WACV57701.2024.00177