{"title":"Formula 1 onboard camera shot detector using motion activity areas","authors":"Arnau Raventos, F. Tarrés","doi":"10.1109/CBMI.2012.6269800","DOIUrl":null,"url":null,"abstract":"Shot detection in sports video sequences has been of great interest in the last years. In this paper, a new approach to detect onboard camera shots in compressed Formula 1 video sequences is presented. To that end, and after studying the characteristics of the shot, a technique based in the thresholding comparison between a high motion area and a stationary one has been devised. Efficient computation is achieved by direct decoding of the motion vectors in the MPEG stream. The shot detection process is done through a frame-by-frame hysteresis thresholding analysis. In order to enhance the results, a SVD shot boundary detector is applied. Promising results are presented that show the validity of the approach.","PeriodicalId":120769,"journal":{"name":"2012 10th International Workshop on Content-Based Multimedia Indexing (CBMI)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 10th International Workshop on Content-Based Multimedia Indexing (CBMI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CBMI.2012.6269800","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Shot detection in sports video sequences has been of great interest in the last years. In this paper, a new approach to detect onboard camera shots in compressed Formula 1 video sequences is presented. To that end, and after studying the characteristics of the shot, a technique based in the thresholding comparison between a high motion area and a stationary one has been devised. Efficient computation is achieved by direct decoding of the motion vectors in the MPEG stream. The shot detection process is done through a frame-by-frame hysteresis thresholding analysis. In order to enhance the results, a SVD shot boundary detector is applied. Promising results are presented that show the validity of the approach.