IDiff-Face: Synthetic-based Face Recognition through Fizzy Identity-Conditioned Diffusion Models

TL;DR

IDiff-Face introduces a novel synthetic face dataset generation method using diffusion models, significantly improving face recognition accuracy and addressing privacy concerns associated with authentic datasets.

Contribution

The paper presents a new conditional diffusion model for generating synthetic face identities with realistic variations, enhancing training data diversity for face recognition.

Findings

Achieved 98.00% accuracy on LFW benchmark.

Outperformed recent synthetic datasets in diversity and discrimination.

Bridged the gap between synthetic and authentic face recognition performance.

Abstract

The availability of large-scale authentic face databases has been crucial to the significant advances made in face recognition research over the past decade. However, legal and ethical concerns led to the recent retraction of many of these databases by their creators, raising questions about the continuity of future face recognition research without one of its key resources. Synthetic datasets have emerged as a promising alternative to privacy-sensitive authentic data for face recognition development. However, recent synthetic datasets that are used to train face recognition models suffer either from limitations in intra-class diversity or cross-class (identity) discrimination, leading to less optimal accuracies, far away from the accuracies achieved by models trained on authentic data. This paper targets this issue by proposing IDiff-Face, a novel approach based on conditional latent diffusion models for synthetic identity generation with realistic identity variations for face recognition training. Through extensive evaluations, our proposed synthetic-based face recognition approach pushed the limits of state-of-the-art performances, achieving, for example, 98.00% accuracy on the Labeled Faces in the Wild (LFW) benchmark, far ahead from the recent synthetic-based face recognition solutions with 95.40% and bridging the gap to authentic-based face recognition with 99.82% accuracy.