Football Penalty Kick Prediction Model Based on Kicker's Pose Estimation

Mauricio Salazar J.A.; Alatrista-Salas H. · ACM International Conference Proceeding Series (2024) · DOI: 10.1145/3674029.3674061

Resumen

This paper describes an innovative methodology for predicting penalty kicks in football based on the pose estimation of the kicker. Our proposal starts with the construction of a corpus of penalty kick videos. Then, we combine semantic video segmentation and 3D pose estimation using the TAM and MMPose methods. For the prediction of the goal area where the kick would land, three deep-learning models were compared, as well as the study of the part of the player's body that affects the prediction task.

Instituciones: Pontificia Universidad Católica del Perú