AutoStats Description
AutoStats provides comprehensive data collection from every basketball game via remote video sources. This greatly expands the data that can be used for performance analysis and scalable scouting efforts. It has already transformed the draft process of one NBA team. AutoStats is an AI-enhanced body detection technology that uses remote video sources to fluidly identify the position and movement of players across the entire court. It does not require in-venue technology. Computer vision has enabled new levels of detail to be extracted from video. Sujoy Ganguly, Stats Perform Director of Computer Vision, discusses the impact of this technology on player-tracking technologies during the MIT Sloan Sports Analytics conference. Player-tracking capabilities have been limited by hardware requirements. AutoStats is able to analyze remote video and provide valuable information about team performance that is not usually available outside of the NBA.