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

2026-03-26

AI Summary

This work proposes a novel analytic data augmentation method for image classification based on moiré interference patterns, addressing the high computational cost and reliance on external data common in existing approaches such as diffusion models or complex feature mixing. By introducing closed-form, multi-scale structured perturbations directly into the training pipeline, the method requires no external data, incurs negligible storage overhead, and achieves extremely low computational cost—synthesizing and discarding each perturbation on-the-fly during training. Integrated with Vision Transformers, the approach significantly outperforms current data-free augmentation techniques on robustness benchmarks including ImageNet-C, ImageNet-R, and adversarial evaluations, generating perturbations for a single image in just 0.0026 seconds.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Mixture of Experts (MoE)
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

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 pipeline without external data. Extensive experiments with Vision Transformers demonstrate that the proposed method consistently improves robustness across multiple benchmarks, including ImageNet-C, ImageNet-R, and adversarial benchmarks, outperforming standard augmentation baselines and existing external-data-free augmentation approaches. These results suggest that analytic interference patterns provide a practical and efficient alternative to data-driven generative augmentation methods.

Problem

• data augmentation
• image classification robustness
• Moire interference
• computational overhead
• external datasets

Innovation

• Moire interference
• procedural augmentation
• closed-form formulation
• data augmentation
• robustness