DiverAge: Reliable Pluralistic Face Aging with Cross-Age Identity Relation Guidance

2026-06-02

AI Summary

Existing face aging methods struggle to simultaneously achieve appearance diversity and cross-age identity consistency. This work proposes DiverAge, a framework based on a diffusion autoencoder that enables diverse generation while introducing CARR—a training-free, inference-time guidance strategy that leverages real-world cross-age identity similarity priors to mitigate identity drift. By integrating stochastic denoising generation, age-conditional semantic modulation, and cross-age identity relationship regulation, the method significantly enhances the temporal (ordinal) coherence of aging sequences while preserving identity fidelity, age accuracy, image quality, and visual diversity.

Abstract

Face aging plays an important role in long-term biometric analysis, cross-age identity verification, and forensic identity analysis. Since the same subject may exhibit multiple plausible appearances at a target age due to genetic, environmental, and lifestyle factors, face aging is inherently a one-to-many generation problem. However, pluralism alone is insufficient for reliable face aging: a model should provide appearance-level candidate diversity within each age group while maintaining sequence-level ordinal reliability across ordered age groups. Existing deterministic aging methods can synthesize visually plausible age-progressed faces, but usually lack stochastic diversity. In contrast, pluralistic aging methods introduce local appearance variations, but often fail to explicitly regulate the identity evolution of the full aging sequence. In this paper, we propose \textbf{DiverAge}, a hierarchical pluralistic face aging framework based on diffusion autoencoding. DiverAge preserves appearance-level diversity through stochastic diffusion decoding and age-conditioned semantic modulation. To improve sequence-level reliability, we introduce a Cross-age Identity Relation Regulator (CARR), an inference-time guidance strategy that jointly denoises multiple target age groups. CARR is guided by a Cross-age Identity Similarity (CIS) prior estimated from real same-identity cross-age pairs, and suppresses excessive cross-age identity drift through one-sided sampling-time guidance without modifying the training objective or introducing extra trainable parameters. Experiments demonstrate that DiverAge improves sequence-level ordinal reliability while maintaining identity preservation, age accuracy, image quality, and appearance-level diversity.

Problem

• face aging
• pluralistic generation
• identity preservation
• sequence-level reliability
• cross-age identity

Innovation

• pluralistic face aging
• diffusion autoencoding
• cross-age identity relation
• sequence-level reliability
• identity preservation