A fast CU size decision algorithm for VVC intra prediction based on support vector machine
- PDF / 1,146,438 Bytes
- 17 Pages / 439.37 x 666.142 pts Page_size
- 35 Downloads / 183 Views
A fast CU size decision algorithm for VVC intra prediction based on support vector machine Fen Chen 1 & Yan Ren 2 & Zongju Peng 1,2
2
& Gangyi Jiang & Xin Cui
2
Received: 28 November 2019 / Revised: 11 May 2020 / Accepted: 21 July 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
The latest generation of coding standard, Versatile Video Coding (VVC), has achieved more bitrate reduction compared with high efficiency video coding. However, the introduction of quadtree with nested Multi-Type Tree (MTT) coding structure greatly increases the computational complexity. To reduce the complexity of VVC, a Support Vector Machine (SVM) based Coding Unit (CU) size decision algorithm is presented. Firstly, effective features, derived from entropy, texture contrast, and Haar wavelet efficient of current CU, are select to distinguish the splitting directions. Then, the six SVM classifying models are on-line trained at different CU sizes. Finally, the models are utilized to prediction the direction of CU splitting in the quadtree with nested MTT coding structure. Experimental results show that the proposed algorithm can significantly save the encoding time by 51.01% with slight increase of Bjontegaard delta bit rate. Keywords Versatile video coding . Fast algorithm . Support vector machine . CU splitting
1 Introduction With the development of digital video technology, especially the emergence of Ultra-High Definition (UHD) video technology, the tremendously increasing data amount of UHD video brings a heavy burden to storage and transmission system. To improve encoding efficiency, ITU-T Video Coding Expert Group (VCEG) and the ISO Motion Picture Expert Group (MPEG) have collaborated to form the Joint Video Experts Team (JEVT), and initiated the exploration of the new generation video coding standard on the base of the High Efficiency Video Coding (HEVC) [21]. In 2015, a transitional standard Future Video Coding (FVC) was
* Zongju Peng [email protected]
1
School of Electrical and Electronic Engineering, Chongqing University of Technology, Chongqing 400054, China
2
Faculty of information Science and Engineering, Ningbo University, Ningbo 315211 Zhejiang, China
Multimedia Tools and Applications
developed [5]. Encouraged by the performance brought by the new tools in FVC, JEVT used the majority of the new tools and established the new generation standard Versatile Video Coding (VVC), in 2018 [3]. The coding efficiency of VVC has improved compared with HEVC. However, the complexity has increased tremendously, which hinders its application. Therefore, it is of great significance to reduce the complexity of VVC. There are a series of new techniques in VVC, such as the new structure of Coding Unit (CU), extended angle modes in intra prediction, affine motion in inter prediction, etc. The most effective but extremely time-consuming technique in VVC is the quadtree with nested Multi-Type Tree (MTT) coding structure. It replaces the quadtree structure in HEVC. In MTT coding structure, the CU
Data Loading...
