Mi-Sook Chung, S. Kang, Jae-Ho Choi, Hoon-Sung Kwak
{"title":"New multiresolution motion estimation for wavelet transform video coding","authors":"Mi-Sook Chung, S. Kang, Jae-Ho Choi, Hoon-Sung Kwak","doi":"10.1109/APCAS.1996.569206","DOIUrl":null,"url":null,"abstract":"In this paper, we propose a new multiresolution motion estimation (NMRME) scheme for wavelet transform based video coding to increase the PSNR. In the NMRME scheme, motion vectors are estimated for low pass subimages of every layers, and all component subimages at the lowest layer by the block matching. At first the motion vectors are calculated for the low pass subimage at highest layer. Then, motion vectors at lower layers are refined using the motion information obtained at higher layers. Motion vectors for vertical, horizontal, and diagonal component subimages at lowest layers are refined using the motion vectors obtained at the low pass subimage of the same layer. Considering the simulation result, the PSNR is higher in the NMRME than that in MRME and FMRME. But, the bit rate is higher. Therefore, tile NMRME scheme is adequate for the application which requires a high PSNR.","PeriodicalId":20507,"journal":{"name":"Proceedings of APCCAS'96 - Asia Pacific Conference on Circuits and Systems","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"1996-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of APCCAS'96 - Asia Pacific Conference on Circuits and Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/APCAS.1996.569206","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
In this paper, we propose a new multiresolution motion estimation (NMRME) scheme for wavelet transform based video coding to increase the PSNR. In the NMRME scheme, motion vectors are estimated for low pass subimages of every layers, and all component subimages at the lowest layer by the block matching. At first the motion vectors are calculated for the low pass subimage at highest layer. Then, motion vectors at lower layers are refined using the motion information obtained at higher layers. Motion vectors for vertical, horizontal, and diagonal component subimages at lowest layers are refined using the motion vectors obtained at the low pass subimage of the same layer. Considering the simulation result, the PSNR is higher in the NMRME than that in MRME and FMRME. But, the bit rate is higher. Therefore, tile NMRME scheme is adequate for the application which requires a high PSNR.