{"title":"An improved parallel architecture fro MPEG-4 motion estimation in 3G mobile applications","authors":"Donglai Xu, Rui Gao, H. Batatia","doi":"10.1109/ICME.2003.1221343","DOIUrl":null,"url":null,"abstract":"A high-parallel VLSI core architecture for MPEG-4 motion estimation is proposed in this paper. It possesses the characteristics of low memory bandwidth and low clock rate requirements, thus primarily aiming at 3G mobile applications. Based on a one-dimensional tree architecture, the architecture employs the dual-register/buffer technique to reduce the preload and alignment cycles. As an example, full-search block matching algorithm has been mapped onto this architecture using a 16-PE array that has the ability to calculate the motion vectors of QCIF video sequences in real time at 1 MHz clock rate and using 15.5 Mbytes/s memory bandwidth.","PeriodicalId":118560,"journal":{"name":"2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2003-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICME.2003.1221343","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
A high-parallel VLSI core architecture for MPEG-4 motion estimation is proposed in this paper. It possesses the characteristics of low memory bandwidth and low clock rate requirements, thus primarily aiming at 3G mobile applications. Based on a one-dimensional tree architecture, the architecture employs the dual-register/buffer technique to reduce the preload and alignment cycles. As an example, full-search block matching algorithm has been mapped onto this architecture using a 16-PE array that has the ability to calculate the motion vectors of QCIF video sequences in real time at 1 MHz clock rate and using 15.5 Mbytes/s memory bandwidth.