Automated Full-Game Ice Hockey Analytics

Computer vision involves creating algorithms capable of interpreting scenes. A key challenge is automatic generation of analytics to mimic human ability. Generating analytics from ice hockey video is one such application where human-captured analytics typically focus only on puck-centric events and it is not feasible for humans to interpret all game events. Computer vision can be developed to interpret all player activities and then use this information to interpret game events, in support of scouting opponents, preparing team game plans, assessing draft selections, and preparing players for games. The challenges involve game speed, persistent occlusions, identifying unique players, and limited video field-of-view. Our published research has made strong progress working with a company, Stathletes, with a successful business enterprise. Our objective is to advance the state-of-the-art in hockey analytics to automatically interpret a broadcast ice hockey video to assist corporate monetization and market penetration.

Harish Prakash;Bavesh Balaji
Faculty Supervisor: 
David Clausi;John Zelek;Alexander Wong;Sirisha Rambhatla;Mohammad Javad Shafiee
Partner University: