Cross Domain Generative Augmentation: Domain Generalization with Latent Diffusion Models

TL;DR

This paper introduces Cross Domain Generative Augmentation (CDGA), a novel data augmentation method using latent diffusion models to generate synthetic images that bridge domain gaps, improving domain generalization performance.

Contribution

The paper proposes a new augmentation technique, CDGA, that estimates data densities between domain pairs with latent diffusion models to enhance domain generalization.

Findings

CDGA outperforms state-of-the-art DG methods on Domainbed benchmark.

Generated over 5 million synthetic images for analysis.

Extensive ablation studies validate the effectiveness of CDGA.

Abstract

Despite the huge effort in developing novel regularizers for Domain Generalization (DG), adding simple data augmentation to the vanilla ERM which is a practical implementation of the Vicinal Risk Minimization principle (VRM) \citep{chapelle2000vicinal} outperforms or stays competitive with many of the proposed regularizers. The VRM reduces the estimation error in ERM by replacing the point-wise kernel estimates with a more precise estimation of true data distribution that reduces the gap between data points \textbf{within each domain}. However, in the DG setting, the estimation error of true data distribution by ERM is mainly caused by the distribution shift \textbf{between domains} which cannot be fully addressed by simple data augmentation techniques within each domain. Inspired by this limitation of VRM, we propose a novel data augmentation named Cross Domain Generative Augmentation (CDGA) that replaces the pointwise kernel estimates in ERM with new density estimates in the \textbf{vicinity of domain pairs} so that the gap between domains is further reduced. To this end, CDGA, which is built upon latent diffusion models (LDM), generates synthetic images to fill the gap between all domains and as a result, reduces the non-iidness. We show that CDGA outperforms SOTA DG methods under the Domainbed benchmark. To explain the effectiveness of CDGA, we generate more than 5 Million synthetic images and perform extensive ablation studies including data scaling laws, distribution visualization, domain shift quantification, adversarial robustness, and loss landscape analysis.