Saved in:
Bibliographic Details
Main Authors: Ranasinghe, Pasindu, Ranasinghe, Pamudu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.26088
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909816643911680
author Ranasinghe, Pasindu
Ranasinghe, Pamudu
author_facet Ranasinghe, Pasindu
Ranasinghe, Pamudu
contents Penalty kicks often decide championships, yet goalkeepers must anticipate the kicker's intent from subtle biomechanical cues within a very short time window. This study introduces a real-time, multi-modal deep learning framework to predict the direction of a penalty kick (left, middle, or right) before ball contact. The model uses a dual-branch architecture: a MobileNetV2-based CNN extracts spatial features from RGB frames, while 2D keypoints are processed by an LSTM network with attention mechanisms. Pose-derived keypoints further guide visual focus toward task-relevant regions. A distance-based thresholding method segments input sequences immediately before ball contact, ensuring consistent input across diverse footage. A custom dataset of 755 penalty kick events was created from real match videos, with frame-level annotations for object detection, shooter keypoints, and final ball placement. The model achieved 89% accuracy on a held-out test set, outperforming visual-only and pose-only baselines by 14-22%. With an inference time of 22 milliseconds, the lightweight and interpretable design makes it suitable for goalkeeper training, tactical analysis, and real-time game analytics.
format Preprint
id arxiv_https___arxiv_org_abs_2509_26088
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Predicting Penalty Kick Direction Using Multi-Modal Deep Learning with Pose-Guided Attention
Ranasinghe, Pasindu
Ranasinghe, Pamudu
Computer Vision and Pattern Recognition
Penalty kicks often decide championships, yet goalkeepers must anticipate the kicker's intent from subtle biomechanical cues within a very short time window. This study introduces a real-time, multi-modal deep learning framework to predict the direction of a penalty kick (left, middle, or right) before ball contact. The model uses a dual-branch architecture: a MobileNetV2-based CNN extracts spatial features from RGB frames, while 2D keypoints are processed by an LSTM network with attention mechanisms. Pose-derived keypoints further guide visual focus toward task-relevant regions. A distance-based thresholding method segments input sequences immediately before ball contact, ensuring consistent input across diverse footage. A custom dataset of 755 penalty kick events was created from real match videos, with frame-level annotations for object detection, shooter keypoints, and final ball placement. The model achieved 89% accuracy on a held-out test set, outperforming visual-only and pose-only baselines by 14-22%. With an inference time of 22 milliseconds, the lightweight and interpretable design makes it suitable for goalkeeper training, tactical analysis, and real-time game analytics.
title Predicting Penalty Kick Direction Using Multi-Modal Deep Learning with Pose-Guided Attention
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.26088