fbpx

Highlights Generation for Tennis Matches using Computer Vision, Natural Language Processing and Audio Analysis

IS Button

Name: Alon Liberman
Major: Computer Science and Mathematics
Advisors: Heather Guarnera and Subhadip Chowdhury

This project uses computer vision, natural language processing and audio analysis to automatize the highlights generation task for tennis matches. Computer vision techniques such as camera shot detection, Hough transform and neural networks are used to extract the time intervals of the points. To detect the best points, three approaches are used. Point length suggests which points correspond to rallies and aces. The audio waves are analyzed to search for the highest audio peaks, which indicate the moments where the crowd cheers the most. Sentiment analysis, a natural language processing technique, is used to look for points where the commentators make positive statements. The software receives a full tennis match with no cuts as the input, and outputs a short summary with the most relevant points. The final software pipeline was tested on three tennis matches from the 2021 US Open for manual validation.

Posted in Comments Enabled, Independent Study, Symposium 2022.


3 responses to “Highlights Generation for Tennis Matches using Computer Vision, Natural Language Processing and Audio Analysis”

  1. Ben Foltz says:

    Alon I’ve known you for all four years, so happy to call you a friend. I could not be more proud of your work! One question for you though, do you see this software being applicable in the world of tennis broadcasting?

  2. saeed says:

    Love how your project has the potential to revolutionize the tennis broadcasting industry with new and innovative ways. Really looking forward to more studies on this from other sports, and ultimately its implementation! Vamos Alon!!

  3. Jillian Morrison says:

    Alon, Great job! This is a clever way to use data that previously has not been used or not used enough. One thing that comes to mind is how much data is being collected (A LOT) and we can think of ways to use all these data to benefit society. Great job doing this! Hopefully you or someone else can continue working on this to turn revolutionize the industry!