Finding your Lookalike: Measuring Face Similarity Rather than Face Identity

TL;DR

This paper introduces a new task of quantifying perceived face similarity, presents a dataset for it, and proposes a specialized neural network that outperforms traditional face recognition models for this purpose.

Contribution

The work defines face similarity as a distinct task from face recognition, creates a dedicated dataset, and develops the Lookalike network for improved similarity prediction.

Findings

Lookalike network outperforms face recognition models on similarity task

Face similarity is a distinct task from face recognition

New dataset for perceived face similarity

Abstract

Face images are one of the main areas of focus for computer vision, receiving on a wide variety of tasks. Although face recognition is probably the most widely researched, many other tasks such as kinship detection, facial expression classification and facial aging have been examined. In this work we propose the new, subjective task of quantifying perceived face similarity between a pair of faces. That is, we predict the perceived similarity between facial images, given that they are not of the same person. Although this task is clearly correlated with face recognition, it is different and therefore justifies a separate investigation. Humans often remark that two persons look alike, even in cases where the persons are not actually confused with one another. In addition, because face similarity is different than traditional image similarity, there are challenges in data collection and labeling, and dealing with diverging subjective opinions between human labelers. We present evidence that finding facial look-alikes and recognizing faces are two distinct tasks. We propose a new dataset for facial similarity and introduce the Lookalike network, directed towards similar face classification, which outperforms the ad hoc usage of a face recognition network directed at the same task.