[2603.25109] MoireMix: A Formula-Based Data Augmentation for Improving Image Classification Robustness

Computer Science > Computer Vision and Pattern Recognition arXiv:2603.25109 (cs) [Submitted on 26 Mar 2026] Title:MoireMix: A Formula-Based Data Augmentation for Improving Image Classification Robustness Authors:Yuto Matsuo, Yoshihiro Fukuhara, Yuki M. Asano, Rintaro Yanagi, Hirokatsu Kataoka, Akio Nakamura View a PDF of the paper titled MoireMix: A Formula-Based Data Augmentation for Improving Image Classification Robustness, by Yuto Matsuo and 4 other authors View PDF HTML (experimental) Abstract:Data augmentation is a key technique for improving the robustness of image classification models. However, many recent approaches rely on diffusion-based synthesis or complex feature mixing strategies, which introduce substantial computational overhead or require external datasets. In this work, we explore a different direction: procedural augmentation based on analytic interference patterns. Unlike conventional augmentation methods that rely on stochastic noise, feature mixing, or generative models, our approach exploits Moire interference to generate structured perturbations spanning a wide range of spatial frequencies. We propose a lightweight augmentation method that procedurally generates Moire textures on-the-fly using a closed-form mathematical formulation. The patterns are synthesized directly in memory with negligible computational cost (0.0026 seconds per image), mixed with training images during training, and immediately discarded, enabling a storage-free augmentation...