# Self-Supervised Learning of Face Representations for Video Face Clustering

## TL;DR

This paper introduces a self-supervised Siamese network for video face clustering that outperforms existing methods by effectively extracting identity information from deep face representations without requiring video-specific supervision.

## Contribution

The paper presents a novel self-supervised learning approach for face clustering that works on both videos and images, improving clustering accuracy without labeled data.

## Key findings

- Outperforms state-of-the-art on three datasets
- Effective without video/track supervision
- Applicable to both video and image collections

## Abstract

Analyzing the story behind TV series and movies often requires understanding who the characters are and what they are doing. With improving deep face models, this may seem like a solved problem. However, as face detectors get better, clustering/identification needs to be revisited to address increasing diversity in facial appearance. In this paper, we address video face clustering using unsupervised methods. Our emphasis is on distilling the essential information, identity, from the representations obtained using deep pre-trained face networks. We propose a self-supervised Siamese network that can be trained without the need for video/track based supervision, and thus can also be applied to image collections. We evaluate our proposed method on three video face clustering datasets. The experiments show that our methods outperform current state-of-the-art methods on all datasets. Video face clustering is lacking a common benchmark as current works are often evaluated with different metrics and/or different sets of face tracks.