Research and Development of an AI based Software to Recognize Realtime TV Commercials on Low-power Computing Equipment

This research will explore the possible use of video analysis, object detection, and machine learning models to identify TV commercials and advertisements in real-time on low-power computer systems. The partner organization focuses on software development and advertisement technology. The partner aims to solve the challenge of developing a lightweight power-efficient AI-based software that would be able to identify TV commercials or other advertisements in live video feeds in real time. This technology will help the company identify and quantify advertisements for tracking purposes. The interns would work on categorizing advertisement data, optimizing the system, and identifying key characteristics associated with TV commercials transitions.

Faculty Supervisor:

Michael Guerzhoy

Student:

Partner:

Smart CashBackTV

Discipline:

Computer science

Sector:

Information and cultural industries

University:

University of Toronto

Program: