Face, Body, Voice: Video Person-Clustering with Multiple Modalities

TL;DR

This paper introduces a multi-modal approach for person-clustering in videos, leveraging face, body, and voice cues, along with a new large dataset, to improve clustering accuracy and support story understanding.

Contribution

It presents a novel multi-modal clustering algorithm, a comprehensive dataset for person-clustering, and demonstrates improved performance and applications in story understanding.

Findings

Achieved state-of-the-art results on face and person-clustering datasets.

Showed multi-modal cues improve clustering accuracy.

Enabled new insights into story understanding through character co-occurrences.

Abstract

The objective of this work is person-clustering in videos -- grouping characters according to their identity. Previous methods focus on the narrower task of face-clustering, and for the most part ignore other cues such as the person's voice, their overall appearance (hair, clothes, posture), and the editing structure of the videos. Similarly, most current datasets evaluate only the task of face-clustering, rather than person-clustering. This limits their applicability to downstream applications such as story understanding which require person-level, rather than only face-level, reasoning. In this paper we make contributions to address both these deficiencies: first, we introduce a Multi-Modal High-Precision Clustering algorithm for person-clustering in videos using cues from several modalities (face, body, and voice). Second, we introduce a Video Person-Clustering dataset, for evaluating multi-modal person-clustering. It contains body-tracks for each annotated character, face-tracks when visible, and voice-tracks when speaking, with their associated features. The dataset is by far the largest of its kind, and covers films and TV-shows representing a wide range of demographics. Finally, we show the effectiveness of using multiple modalities for person-clustering, explore the use of this new broad task for story understanding through character co-occurrences, and achieve a new state of the art on all available datasets for face and person-clustering.