Research Sports Video Analysis
Play analysis for rugby video by using time series Deep Learning and multimodal information
For the purpose of constructing a sports image analysis system using image recognition technology, we designed a system based on the concept that you can look back at the desired scene while understanding the game situation in a bird's-eye view. We focused on rugby and worked on players and ball tracking technologies from actual game videos requested by specialist analysts. In addition, we developed a technique to estimate the position of the ball and each player when mapping it on the 2D field, and prototyped a demonstration system assuming game analysis. Furthermore, we developed a technique for play classification by time series Deep Learning.