{"title":"Texture-based fast inter mode selection algorithm for H.264","authors":"Likang Kou, N. Yu, Meihua Gu","doi":"10.1109/KAM.2010.5646173","DOIUrl":null,"url":null,"abstract":"In this paper, a fast inter-frame mode selection algorithm based on texture is proposed. The algorithm uses three model sets, the macro block level, the sub-macroblock level and the block level to divide all 7 inter modes. There are three methods of texture analysis to determine the best reference mode on the each level. The final decision is made by calculating the RDO-cost of the three optimal level models and choosing one with minimum cost. Experimental results comparing the proposed method with original show that this fast algorithm is more advanced and effective. The average encoding time is reduced by 60%, in some cases, by 70%, while PSNR and encoding bits are affected less.","PeriodicalId":160788,"journal":{"name":"2010 Third International Symposium on Knowledge Acquisition and Modeling","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2010 Third International Symposium on Knowledge Acquisition and Modeling","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/KAM.2010.5646173","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
In this paper, a fast inter-frame mode selection algorithm based on texture is proposed. The algorithm uses three model sets, the macro block level, the sub-macroblock level and the block level to divide all 7 inter modes. There are three methods of texture analysis to determine the best reference mode on the each level. The final decision is made by calculating the RDO-cost of the three optimal level models and choosing one with minimum cost. Experimental results comparing the proposed method with original show that this fast algorithm is more advanced and effective. The average encoding time is reduced by 60%, in some cases, by 70%, while PSNR and encoding bits are affected less.