Sports highlights generation based on acoustic events detection: A rugby case study

TL;DR

This paper presents a system that generates sports highlights from broadcasts by detecting key acoustic events using a language-independent, multi-stage classification approach, demonstrating high efficiency through objective and human evaluations.

Contribution

The study introduces a novel audio-based highlight generation method for sports, specifically rugby, utilizing a multi-stage acoustic event detection framework.

Findings

High efficiency in highlight detection

Effective language-independent acoustic classification

Positive human evaluation results

Abstract

We approach the challenging problem of generating highlights from sports broadcasts utilizing audio information only. A language-independent, multi-stage classification approach is employed for detection of key acoustic events which then act as a platform for summarization of highlight scenes. Objective results and human experience indicate that our system is highly efficient.