Automated multi-target tracking in broadcast video cameras

Automated object detection and tracking in videos is still one of the challenging problems in computer vision. Generally, it is a very challenging problem due to the loss of information caused by the projection of the 3D world on a 2D image, noise in images, cluttered background, complex object motion, partial or full occlusions, changes in illumination, real-time processing requirements, etc. This project is about detection, identification, and estimation the location of all hockey players in a game in the real world co-ordinates using the broadcast video cameras. Therefore, this project contributes to constructing a fully automated and robust multi-object tracking system for sport player tracking that works in real-time with respect to the current hardware resources.

Faculty Supervisor:

Gregory Dudek

Student:

Juan Camilo Gamboa Higuera

Partner:

SPORTLOGiQ Inc.

Discipline:

Computer science

Sector:

Sports and recreation

University:

McGill University