Fast Adaptive CU Partition Decision Algorithm for VVC Intra Coding
Open Access
- 25 October 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 119766-119778
Abstract
The latest video standard - Versatile Video Coding Standard (VVC/H.266) has been standardized and officially entered into force. Compared with the High Efficiency Video Coding (HEVC/H.265), owing to the introduction of the Quad-tree with Nested Multi-type Tree (QTMT) division mode, the encoder can choose a more detailed division type when dividing the Coding unit (CU), thereby improving the coding performance. However, when the CU selects the division type, it needs to traverse all possible division types and compute the Rate-distortion (RD) cost, which greatly enhances the coding complexity. Therefore, this paper designs a Joint Random Forest Classifier (JRFC) to make decisions on CU partition types, and proposes fast adaptive CU partition decision algorithm for VVC intra coding combined with our previous work. The algorithm has the capability to make partition decisions for CUs of diverse sizes (smaller than 32x32) and completely bypass the intricate process of Rate-distortion Optimization (RDO), resulting in a significant reduction in encoding time. The experimental results demonstrate that, in comparison to the basic approach utilized in VTM10.0, the algorithm proposed in this paper reduces the encoding time by 57.27% on average, with only a marginal increase of 1.53% in Bjontegaard delta bit rate (BDBR).Keywords
Funding Information
- National Natural Science Foundation of China (61771432, 61302118)
- Basic Research Projects of Education Department of Henan (21zx003, 20A880004)
- Key Projects of Natural Science Foundation of Henan (232300421150)
- Scientific and Technological Project of Henan Province (232102211014)
- Postgraduate Education Reform and Quality Improvement Project of Henan Province (YJS2021KC12, YJS2023JC08, YJS2022AL034)
This publication has 24 references indexed in Scilit:
- Tunable VVC Frame Partitioning Based on Lightweight Machine LearningIEEE Transactions on Image Processing, 2019
- Fast Partitioning Decision Strategies for The Upcoming Versatile Video Coding (VVC) Standard2019
- A CNN-based Coding Unit Partition in HEVC for Video Processing2019
- Fast CU Partitioning Algorithm for H.266/VVC Intra-Frame Coding2019
- Random Forest Oriented Fast QTBT Frame Partitioning2019
- A Fast CTU Depth Selection Algorithm for H.265/HEVC Based on Machine Learning2018
- Fast Algorithm for HEVC Intra Prediction Based on Adaptive Mode Decision and Early Termination of CU Partition2018
- Computational complexity reduction for HEVC intra prediction with SVM2017
- Machine Learning-Based Coding Unit Depth Decisions for Flexible Complexity Allocation in High Efficiency Video CodingIEEE Transactions on Image Processing, 2015
- Comparison of the Coding Efficiency of Video Coding Standards—Including High Efficiency Video Coding (HEVC)IEEE Transactions on Circuits and Systems for Video Technology, 2012