Low computation method for fast mode decision in H.264/MPEG-4 AVC

S. Tran, M. Preda, F. Prêteux
{"title":"Low computation method for fast mode decision in H.264/MPEG-4 AVC","authors":"S. Tran, M. Preda, F. Prêteux","doi":"10.1109/IWSSIP.2008.4604471","DOIUrl":null,"url":null,"abstract":"The efficiency of the H.264/AVC video coding standard (the latest in the MPEG family) can be further improved by advancing new prediction key-techniques. However, generally these techniques also lead to an extremely high complexity encoding process. As a solution to this problem, we focused on a fast technique to avoid the exhaustive search on the numerous prediction methods available in the standard. Exploiting the high spatial correlation in the natural video, we introduce a new measure of conformance (MC), which is based on the similarity of performance produced by the encoder on the neighbour blocks. This MC makes it possible to select automatically the sub-optimal prediction mode for intra coding as well as to safely avoid the full-scale search for motion compensation applied to variable block-size throughout multiple reference pictures. The MC is updated during the encoding process of the considered block. Therefore, it does not expose any serious additional computation. The experimental results obtained by the new technique are discussed in the paper. In average, the H.264 encoder is speeded up by 23 % at a slight decrease in compression efficiency (5.8 %) for all most the same PSNR quality (-0.08 %).","PeriodicalId":322045,"journal":{"name":"2008 15th International Conference on Systems, Signals and Image Processing","volume":"53 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 15th International Conference on Systems, Signals and Image Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IWSSIP.2008.4604471","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

The efficiency of the H.264/AVC video coding standard (the latest in the MPEG family) can be further improved by advancing new prediction key-techniques. However, generally these techniques also lead to an extremely high complexity encoding process. As a solution to this problem, we focused on a fast technique to avoid the exhaustive search on the numerous prediction methods available in the standard. Exploiting the high spatial correlation in the natural video, we introduce a new measure of conformance (MC), which is based on the similarity of performance produced by the encoder on the neighbour blocks. This MC makes it possible to select automatically the sub-optimal prediction mode for intra coding as well as to safely avoid the full-scale search for motion compensation applied to variable block-size throughout multiple reference pictures. The MC is updated during the encoding process of the considered block. Therefore, it does not expose any serious additional computation. The experimental results obtained by the new technique are discussed in the paper. In average, the H.264 encoder is speeded up by 23 % at a slight decrease in compression efficiency (5.8 %) for all most the same PSNR quality (-0.08 %).
H.264/MPEG-4 AVC中快速模式判定的低计算方法
H.264/AVC视频编码标准(MPEG家族中最新的一种)可以通过提出新的预测关键技术来进一步提高编码效率。然而,通常这些技术也导致了极高的编码复杂度。为了解决这一问题,我们重点研究了一种快速的技术,以避免对标准中可用的众多预测方法进行穷举搜索。利用自然视频中的高空间相关性,我们引入了一种新的一致性度量(MC),该度量基于编码器在相邻块上产生的性能相似性。该MC可以自动选择帧内编码的次优预测模式,并且可以安全地避免在多个参考图像中应用于可变块大小的运动补偿的全面搜索。在考虑的块的编码过程中更新MC。因此,它不会暴露任何严重的额外计算。文中讨论了新技术的实验结果。平均而言,在几乎相同的PSNR质量(- 0.08%)下,H.264编码器在压缩效率(5.8%)略有下降的情况下加速了23%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信