1 / 1

Automating Game Event Classification in Ice Hockey Through Video Analysis

This project focuses on automating the identification of game events in Union College men’s ice hockey films. By manually gathering player location and orientation attributes, we classify game events from still video frames. Our proposed system will enhance video analysis for coaching, reliably recognizing key actions like breakouts, face-offs, and scoring chances. Initial results show high classification accuracy, with location attributes proving vital. Future goals include automating data gathering through object tracking and recognizing additional game events.

yaakov
Download Presentation

Automating Game Event Classification in Ice Hockey Through Video Analysis

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Senior Project – Computer Science – 2013Classification of Game Eventsin Ice Hockey Game Film Phil CohnAdvisor – Prof. Cass Abstract In this project we aim to automate a way to identify what is happening in Union College men’s ice hockey game film. We manually gather location and orientation attributes for each individual player and use this information to classify game events in still video frames. This classification system aims to play a primary role in a comprehensive video analysis system capable of recognizing and classifying ice hockey game events automatically for coaching purposes. Attributes Breakout Face-off Scoring Chance Results • Classification accuracy is high • Grid attribute is more useful for location than zone attribute • Location is the most important attribute Future Work We aim to expand our system to gather player attribute information automatically using object tracking techniques and action recognition techniques. Additionally we aim to add additional game events for classification.

More Related