Unbiased Image Synthesis via Manifold Guidance in Diffusion Models

arXiv:2307.08199v3 Announce Type: replace
Abstract: Diffusion Models are a potent class of generative models capable of producing high-quality images. However, they often inadvertently favor certain data attributes, undermining the diversity of generated images. This issue is starkly apparent in skewed datasets like CelebA, where the initial dataset disproportionately favors females over males by 57.9%, this bias amplified in generated data where female representation outstrips males by 148%. In response, we propose a plug-and-play method named Manifold Guidance Sampling, which is …

abstract arxiv class cs.cv data dataset datasets diffusion diffusion models diversity generated generative generative models guidance however image images issue manifold quality synthesis type unbiased via