{"title":"A performance evaluation framework for video stabilization methods","authors":"Wilko Guilluy, Azeddine Beghdadi, L. Oudre","doi":"10.1109/EUVIP.2018.8611729","DOIUrl":null,"url":null,"abstract":"This study aims at discussing both objective and subjective aspects of Video Stabilization Quality Assessment (VSQA). A corpus of various degraded videos representing different challenging scenarios and their corresponding processed outputs obtained by four representative methods of Video Stabilization techniques is dedicated to this study. The objective evaluation is restricted to four common VSQA metrics and a new one. The subjective experiments were performed in a laboratory controlled environment using pair wise comparison ranking protocol. Through the obtained results it is shown that the performance evaluation of VS methods is far from being understood and there is still a way to go before satisfactory approaches emerge. This contribution is a first step into the direction of filling this gap by proposing a video stabilization quality assessment methodology.","PeriodicalId":252212,"journal":{"name":"2018 7th European Workshop on Visual Information Processing (EUVIP)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 7th European Workshop on Visual Information Processing (EUVIP)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/EUVIP.2018.8611729","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
Abstract
This study aims at discussing both objective and subjective aspects of Video Stabilization Quality Assessment (VSQA). A corpus of various degraded videos representing different challenging scenarios and their corresponding processed outputs obtained by four representative methods of Video Stabilization techniques is dedicated to this study. The objective evaluation is restricted to four common VSQA metrics and a new one. The subjective experiments were performed in a laboratory controlled environment using pair wise comparison ranking protocol. Through the obtained results it is shown that the performance evaluation of VS methods is far from being understood and there is still a way to go before satisfactory approaches emerge. This contribution is a first step into the direction of filling this gap by proposing a video stabilization quality assessment methodology.