MoireDB: Formula-generated Interference-fringe Image Dataset

TL;DR

MoireDB is a novel, formula-generated interference-fringe image dataset designed to improve image recognition robustness, addressing issues of copyright, cost, and scalability inherent in previous augmentation methods.

Contribution

It introduces a scalable, copyright-free dataset for image augmentation that enhances model robustness against real-world degradations.

Findings

MoireDB-augmented images outperform fractal arts and FVis-based methods.

The dataset reduces costs and copyright concerns.

Enhanced robustness demonstrated in experiments.

Abstract

Image recognition models have struggled to treat recognition robustness to real-world degradations. In this context, data augmentation methods like PixMix improve robustness but rely on generative arts and feature visualizations (FVis), which have copyright, drawing cost, and scalability issues. We propose MoireDB, a formula-generated interference-fringe image dataset for image augmentation enhancing robustness. MoireDB eliminates copyright concerns, reduces dataset assembly costs, and enhances robustness by leveraging illusory patterns. Experiments show that MoireDB augmented images outperforms traditional Fractal arts and FVis-based augmentations, making it a scalable and effective solution for improving model robustness against real-world degradations.