{"title":"Subjective-quality-based MPEG-2 video compression","authors":"Cheng-Yu Pai, W. Lynch","doi":"10.1109/ITCC.2002.1000425","DOIUrl":null,"url":null,"abstract":"Traditional MPEG-2 video compression uses test model 5 (TM5) as the rate-control algorithm, which does not consider the perceptual quality of the compressed sequence. This paper proposes a new rate-control algorithm, which integrates TM5 with the Watson's digital-video-quality (DVQ) metric. As a proponent of the VQEG (Video Quality Expert Group), the Watson's DVQ is a subjective quality metric based on the human visual system (HVS) which estimates the perceptual quality of human viewers. Simulation results indicate that the proposed algorithm outperforms TM5 for low-motion video sequence, and performs about the same as TM5 for high-motion video sequences.","PeriodicalId":115190,"journal":{"name":"Proceedings. International Conference on Information Technology: Coding and Computing","volume":"84 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2002-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings. International Conference on Information Technology: Coding and Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ITCC.2002.1000425","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Traditional MPEG-2 video compression uses test model 5 (TM5) as the rate-control algorithm, which does not consider the perceptual quality of the compressed sequence. This paper proposes a new rate-control algorithm, which integrates TM5 with the Watson's digital-video-quality (DVQ) metric. As a proponent of the VQEG (Video Quality Expert Group), the Watson's DVQ is a subjective quality metric based on the human visual system (HVS) which estimates the perceptual quality of human viewers. Simulation results indicate that the proposed algorithm outperforms TM5 for low-motion video sequence, and performs about the same as TM5 for high-motion video sequences.