Takahiro Itazuri, Tsukasa Fukusato, Shugo Yamaguchi, Shigeo Morishima

Court-based Volleyball Video Summarization Focusing on Rally Scene

DOI: 10.1109/CVPRW.2017.28

Computer Vision and Pattern Recognition Workshop

http://ieeexplore.ieee.org/document/8014762/

Download Sakuranbo. Our system generates a summary video that consists of only more important rally scenes evaluated by rally-rank. To reflects viewers' preference more, features indicating the contents of the game should be necessary; however such features have not been considered in most of previous methods nprotect online security 다운로드. Although several visual features such as the position of a ball and players should be used, acquisition of such features in still not robust and unreliable in low-resolution or low frame rate volleyball videos 중국어번역기. Instead, we use the court transition information caused by camera operation. Experimental results demonstrate the effectiveness that our method reflects viewers' preferences over previous methods by evaluating the adjusted R-squared values between mean subjective values and evaluated rally-rank values."}" data-sheets-userformat="{"2":513,"3":[null,0],"12":0}">In this paper, we propose a rally-rank evaluation based on the court transition information for automatic volleyball video summarization considering the contents of the game 더봄 무비. Our system generates a summary video that consists of only more important rally scenes evaluated by rally-rank. To reflects viewers’ preference more, features indicating the contents of the game should be necessary; however such features have not been considered in most of previous methods Download weather nationwide. Although several visual features such as the position of a ball and players should be used, acquisition of such features in still not robust and unreliable in low-resolution or low frame rate volleyball videos 잡코리아 이력서. Instead, we use the court transition information caused by camera operation. Experimental results demonstrate the effectiveness that our method reflects viewers’ preferences over previous methods by evaluating the adjusted R-squared values between mean subjective values and evaluated rally-rank values Stata 14 free.