文献类型：期刊文献

英文题名：Bridging the Gap in Facial Age Progression: An Attention Mechanism Approach

作者：Liu, Taoli[1];Liang, Yubin[1];Wu, Wenchen[1];Tang, Yize[1]

机构：[1]Guangdong Ocean Univ, Fac Math & Comp Sci, Zhanjiang 524088, Peoples R China

年份：2024

卷号：12

起止页码：163682

外文期刊名：IEEE ACCESS

收录：SCI-EXPANDED(收录号：WOS:001354547600001)、、EI(收录号：20244517338794)、Scopus(收录号：2-s2.0-85208370060)、WOS

基金：This work was supported in part by the Guangdong Ocean University Education Reform Project on Blended Teaching of "Computer Organization and Structure" under Grant 580320087, in part by the Guangdong Ocean University Quality Engineering Project: Industry-Education Integration Practice Teaching Base for Software Engineering with Chinasoft International under Grant PX-129223525, in part by the Guangdong Intelligence Platform of Prawn Modern Seed Industry under Project 2022GCZX001, in part by the Research and Application Demonstration of Key Technologies for Intelligent Prawn Breeding under Project 2023ZDZX4012, and in part by the Construction and Industrial Application Demonstration of the Intelligent Platform for Modern Prawn Seed Industry.

语种：英文

外文关键词：Aging; Skin; Training; Predictive models; Attention mechanisms; Image color analysis; Facial features; Diffusion models; Feature extraction; Noise measurement; Face detection; Facial aging prediction; attention mechanism; image generation; computer vision

外文摘要：With the advent of Generative Adversarial Networks (GANs), significant progress has been made in facial aging prediction. However, existing methods still face considerable challenges. Many studies estimate the ages of individuals in images based on their birth dates rather than visual cues, leading to discrepancies between the predicted ages and the actual appearance of facial aging. Moreover, these approaches often overlook racial consistency, resulting in models predominantly tailored to European populations, which limits their generalizability across different races. To address these issues, we propose a novel facial aging prediction framework that employs three independent encoders to model identity, texture features, and facial skeletal structure. We replace traditional convolutional networks with an attention mechanism-based backbone, integrating spatial and channel attention mechanisms to capture both spatial relationships and age-related feature importance. These attention-enhanced feature maps are then processed through a pyramid feature fusion architecture to facilitate multi-scale feature extraction. Our model effectively captures the subtleties of facial aging across different demographics. Extensive experiments and ablation studies demonstrate that our approach excels in preserving identity, ensuring racial consistency, and generating realistic aging effects. The results further highlight the superior ability of the attention mechanisms to extract detailed, localized features essential for facial aging prediction.