Look Across Elapse: Disentangled Representation Learning and Photorealistic Cross-Age Face Synthesis for Age-Invariant Face Recognition

2 Sep 2018Jian ZhaoYu ChengYi ChengYang YangHaochong LanFang ZhaoLin XiongYan XuJianshu LiSugiri PranataShengmei ShenJunliang XingHengzhu LiuShuicheng YanJiashi Feng

Despite the remarkable progress in face recognition related technologies, reliably recognizing faces across ages still remains a big challenge. The appearance of a human face changes substantially over time, resulting in significant intra-class variations... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Age-Invariant Face Recognition CACDVS AIM Accuracy 99.38% # 2
Age-Invariant Face Recognition CACDVS AIM + CAFR Accuracy 99.76% # 1
Age-Invariant Face Recognition CAFR AIM Accuracy 84.81% # 1
Age-Invariant Face Recognition FG-NET AIM Accuracy 93.20% # 1
Face Verification IJB-C AIM TAR @ FAR=0.01 93.50% # 1
Age-Invariant Face Recognition MORPH Album2 AIM + CAFR Rank-1 Recognition Rate 99.65% # 1
Age-Invariant Face Recognition MORPH Album2 AIM Rank-1 Recognition Rate 99.13% # 2