# Dual-reference Face Retrieval

## TL;DR

This paper introduces a dual reference face retrieval system that uses two images to retrieve faces matching a specific identity and age, addressing a novel challenge in face retrieval.

## Contribution

It proposes a joint manifold and quartet-based model to effectively combine identity and age similarities for face retrieval.

## Key findings

- Outperforms hierarchical methods in experiments.
- Successfully retrieves faces at specific ages without explicit age labels.
- Demonstrates the effectiveness of joint manifold and dual-reference approach.

## Abstract

Face retrieval has received much attention over the past few decades, and many efforts have been made in retrieving face images against pose, illumination, and expression variations. However, the conventional works fail to meet the requirements of a potential and novel task --- retrieving a person's face image at a specific age, especially when the specific 'age' is not given as a numeral, i.e. 'retrieving someone's image at the similar age period shown by another person's image'. To tackle this problem, we propose a dual reference face retrieval framework in this paper, where the system takes two inputs: an identity reference image which indicates the target identity and an age reference image which reflects the target age. In our framework, the raw images are first projected on a joint manifold, which preserves both the age and identity locality. Then two similarity metrics of age and identity are exploited and optimized by utilizing our proposed quartet-based model. The experiments show promising results, outperforming hierarchical methods.