# Video Face Clustering with Unknown Number of Clusters

## TL;DR

This paper introduces Ball Cluster Learning (BCL), a supervised method for face clustering in videos with unknown numbers of characters, including minor and background characters, by partitioning embedding space into equal-sized balls.

## Contribution

The paper proposes BCL, a novel supervised clustering approach that estimates the number of clusters and handles challenging real-world video face clustering scenarios.

## Key findings

- BCL effectively estimates the number of clusters in face tracking data.
- The method achieves promising results on standard datasets.
- It adapts metric learning techniques for improved clustering accuracy.

## Abstract

Understanding videos such as TV series and movies requires analyzing who the characters are and what they are doing. We address the challenging problem of clustering face tracks based on their identity. Different from previous work in this area, we choose to operate in a realistic and difficult setting where: (i) the number of characters is not known a priori; and (ii) face tracks belonging to minor or background characters are not discarded. To this end, we propose Ball Cluster Learning (BCL), a supervised approach to carve the embedding space into balls of equal size, one for each cluster. The learned ball radius is easily translated to a stopping criterion for iterative merging algorithms. This gives BCL the ability to estimate the number of clusters as well as their assignment, achieving promising results on commonly used datasets. We also present a thorough discussion of how existing metric learning literature can be adapted for this task.