Fast Hevc Intra Coding Algorithm Based On Machine Learning And Laplacian Transparent Composite Model
2017 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)(2017)
摘要
Compared with H.264, High Efficient Video Coding (HEVC) improves the coding efficiency by 50% at the price of significant increase in encoding time, due to Rate Distortion Optimization (RDO) on large variations of block sizes and prediction modes. In this paper, a fast intra coding algorithm is proposed to alleviate the high computational complexity of HEVC intra-frame coding. The proposed algorithm is based on machine learning and Laplacian Transparent Composite Model (LPTCM). Features called Summation of Binarized Outlier Coefficient (SBOC) vectors are firstly extracted from original frames by using LPTCM and then fed into online trained Support Vector Machine (SVM). Two SVMs are combined to predict Coding Unit (CU) decisions so that the encoding process can be significantly sped up. Additionally, a performance controller is introduced to ensure the robustness of machine learning models. It is shown by experiments that compared with HM 16.3, the proposed algorithm reduces the encoding time, on average, by 48% with negligible increase in BD-rate.
更多查看译文
关键词
High Efficiency Video Coding, Fast Intra Coding, Support Vector Machine, Transparent Composite Model
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络