Content-Aware Video Analysis of Sport Videos using Machine Learning
Abstract - The games content examination has expanded quickly in ongoing many years, in view of the such tremendous development of video transmission over the Web and the interest for computerized telecom applications. In which enormous information produced by sport recordings so there is testing errand of mining or characterization of game video information spreading over web. Existing studies have zeroed in on the techniques of sports video examination from the spatiotemporal perspective (Meta information) rather than a substance based viewpoint(actions). Rationale behind proposed work is to parse sport recordings in light of activities only which sort of game is playing in current video grouping rather than crude data about that video. The game video broadcast is the vitally satisfied spreading and requesting over web. The monstrous interest for sports video broadcasting, many game applications, for example, hot-star, you-tube and numerous webbased entertainment applications. Ongoing many years the games programs have turned into a prevailing concentration in the field of diversion. Research on big data analytics has attracted much attention to machine learning and artificial intelligence techniques. Thus, there is need major areas of strength for of which can deal with such colossal game information in light of content of videos.
Keywords - Action Recognition, Content Aware System, Content-based Multimedia Analysis, Event Detection, Semantic Analysis, Sports, Survey.