快速HEVC畫面間預測編碼之研究;Research on Fast HEVC Inter Prediction Coding

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Communication Engineering > Electronic Thesis & Dissertation > Item 987654321/79556

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/79556

Title:	快速HEVC畫面間預測編碼之研究;Research on Fast HEVC Inter Prediction Coding
Authors:	蔡易達;Tsai, Yi-Ta
Contributors:	通訊工程學系
Keywords:	HEVC;畫面間預測;編碼單元決策;預測模式決策
Date:	2019-01-25
Issue Date:	2019-04-02 15:01:01 (UTC+8)
Publisher:	國立中央大學
Abstract:	近年來發展出越來越多支援高解析度的多媒體產品，為了能夠有效壓縮伴隨高解析度視訊檔案而來的龐大資料量，HEVC使用許多技術來進行編碼得以降低位元率，但卻造成了編碼複雜度大量的提升。為了降低畫面間預測的編碼複雜度，本論文利用最佳預測模式的大小、合併模式的CBF和當前編碼區塊與參考畫面的重建區塊經進階型移動向量預測(AMVP)之起始搜尋點做逐點相減取絕對值之和(SAD)來提早決策編碼單元的分塊，以及利用像素值變異數來判斷區塊運動特性，針對移動緩慢的區塊，利用跳過模式的RDcost；針對移動快速的區塊，則利用加權RDcost閥值、畫面間2Nx2N的CBF以及運動向量差在預測單元畫面間2Nx2N模式做是否提前終止的決策，來節省編碼的時間。在最後我們將其結合，由實驗結果得知，與HEVC標準模式相比，整體平均的BDBR上升0.4%，編碼時間則節省了46.58%。;In recent years, a lot of high resolution multimedia devices have been developed. In order to compress the significant data effectively, HEVC utilize multiple techniques to decrease bitrate efficiently, it also caused a large coding complexity. To reduce coding complexity of inter prediction, this paper uses the size of best prediction mode, Coded Block Flag(CBF) of Merge mode and the sum of absolute values (SAD) of the current coding block and the reconstructed block of the reference picture by Adaptive Motion Vector Prediction(AMVP) to early terminate the segmentation of Coding Unit(CU). We also use pixel variance to decide the block motion characteristics. For slow motion block, use RDcost of skip mode；for fast motion block, use weighted RDcost threshold, Coded Block Flag(CBF) of Inter2Nx2N mode and motion vector difference to early terminate the prediction mode decision in Inter2Nx2N mode. Finally, we combine these two algorithm. According to experiment results, Compare with HEVC, our proposed algorithm can reduced encoding time by 46.58% and BDBR only increased by 0.4%.
Appears in Collections:	[Graduate Institute of Communication Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	157	View/Open

社群 sharing

Loading...