AgeBooth: Controllable Facial Aging and Rejuvenation via Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Shihao, Cao, Bohan, Ouyang, Ziheng, Li, Zhen, Jiang, Peng-Tao, Hou, Qibin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908579275997184
author Zhu, Shihao
Cao, Bohan
Ouyang, Ziheng
Li, Zhen
Jiang, Peng-Tao
Hou, Qibin
author_facet Zhu, Shihao
Cao, Bohan
Ouyang, Ziheng
Li, Zhen
Jiang, Peng-Tao
Hou, Qibin
contents Recent diffusion model research focuses on generating identity-consistent images from a reference photo, but they struggle to accurately control age while preserving identity, and fine-tuning such models often requires costly paired images across ages. In this paper, we propose AgeBooth, a novel age-specific finetuning approach that can effectively enhance the age control capability of adapterbased identity personalization models without the need for expensive age-varied datasets. To reduce dependence on a large amount of age-labeled data, we exploit the linear nature of aging by introducing age-conditioned prompt blending and an age-specific LoRA fusion strategy that leverages SVDMix, a matrix fusion technique. These techniques enable high-quality generation of intermediate-age portraits. Our AgeBooth produces realistic and identity-consistent face images across different ages from a single reference image. Experiments show that AgeBooth achieves superior age control and visual quality compared to previous state-of-the-art editing-based methods.
format Preprint
id arxiv_https___arxiv_org_abs_2510_05715
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AgeBooth: Controllable Facial Aging and Rejuvenation via Diffusion Models
Zhu, Shihao
Cao, Bohan
Ouyang, Ziheng
Li, Zhen
Jiang, Peng-Tao
Hou, Qibin
Computer Vision and Pattern Recognition
Recent diffusion model research focuses on generating identity-consistent images from a reference photo, but they struggle to accurately control age while preserving identity, and fine-tuning such models often requires costly paired images across ages. In this paper, we propose AgeBooth, a novel age-specific finetuning approach that can effectively enhance the age control capability of adapterbased identity personalization models without the need for expensive age-varied datasets. To reduce dependence on a large amount of age-labeled data, we exploit the linear nature of aging by introducing age-conditioned prompt blending and an age-specific LoRA fusion strategy that leverages SVDMix, a matrix fusion technique. These techniques enable high-quality generation of intermediate-age portraits. Our AgeBooth produces realistic and identity-consistent face images across different ages from a single reference image. Experiments show that AgeBooth achieves superior age control and visual quality compared to previous state-of-the-art editing-based methods.
title AgeBooth: Controllable Facial Aging and Rejuvenation via Diffusion Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.05715