Learning from the Pros: Extracting Professional Goalkeeper Technique from Broadcast Footage

TL;DR

This paper develops a computer vision and machine learning framework to analyze professional goalkeeper techniques from broadcast footage, aiming to help amateur goalkeepers learn effective skills by identifying optimal techniques in various match situations.

Contribution

It introduces an unsupervised learning approach using 3D body pose data to evaluate and extract professional goalkeeper techniques from broadcast footage.

Findings

Successful extraction of professional goalkeeper techniques

Development of an 'expected saves' model for optimal technique identification

Framework applicable to different match contexts

Abstract

As an amateur goalkeeper playing grassroots soccer, who better to learn from than top professional goalkeepers? In this paper, we harness computer vision and machine learning models to appraise the save technique of professionals in a way those at lower levels can learn from. We train an unsupervised machine learning model using 3D body pose data extracted from broadcast footage to learn professional goalkeeper technique. Then, an "expected saves" model is developed, from which we can identify the optimal goalkeeper technique in different match contexts.