Age Prediction From Face Images Via Contrastive Learning

TL;DR

This paper introduces a contrastive learning approach for age prediction from face images that effectively extracts age-related features, achieving state-of-the-art results without needing longitudinal data of the same individuals.

Contribution

The paper proposes a novel contrastive learning method that isolates age-related features from face images, overcoming data collection challenges and improving prediction accuracy.

Findings

Achieved state-of-the-art performance on FG-NET dataset.

Outperformed existing methods on MORPH-II dataset.

Effectively suppresses identity features while emphasizing age features.

Abstract

This paper presents a novel approach for accurately estimating age from face images, which overcomes the challenge of collecting a large dataset of individuals with the same identity at different ages. Instead, we leverage readily available face datasets of different people at different ages and aim to extract age-related features using contrastive learning. Our method emphasizes these relevant features while suppressing identity-related features using a combination of cosine similarity and triplet margin losses. We demonstrate the effectiveness of our proposed approach by achieving state-of-the-art performance on two public datasets, FG-NET and MORPH-II.