Fast Inter Prediction Mode Decision Algorithm Based on Data Mining

Tengrui Shi; Xiaobo Guo; Daihui Mo; Jian Wang

Machine Learning and Intelligent Communications. Third International Conference, MLICOM 2018, Hangzhou, China, July 6-8, 2018, Proceedings

Research Article

Fast Inter Prediction Mode Decision Algorithm Based on Data Mining

Download

117 downloads

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1007/978-3-030-00557-3_10,
    author={Tengrui Shi and Xiaobo Guo and Daihui Mo and Jian Wang},
    title={Fast Inter Prediction Mode Decision Algorithm Based on Data Mining},
    proceedings={Machine Learning and Intelligent Communications. Third International Conference, MLICOM 2018, Hangzhou, China, July 6-8, 2018, Proceedings},
    proceedings_a={MLICOM},
    year={2018},
    month={10},
    keywords={HEVC Inter prediction Data mining Decision trees},
    doi={10.1007/978-3-030-00557-3_10}
}

Tengrui Shi
Xiaobo Guo
Daihui Mo
Jian Wang
Year: 2018
Fast Inter Prediction Mode Decision Algorithm Based on Data Mining
MLICOM
Springer
DOI: 10.1007/978-3-030-00557-3_10

Tengrui Shi, Xiaobo Guo¹, Daihui Mo, Jian Wang²^,*

1: The 54th Institute of CETC
2: Nanjing University, NJU

*Contact email: wangjnju@nju.edu.cn

Abstract

The HEVC greatly improves coding efficiency. However, this is accompanied by an increase in the complexity of the coding calculation, which is higher than H.264. We find that there are several features that are highly correlated with the CU’s best split decision in inter prediction. As a result, we choose decision trees to solve the splitting decision problem. We implement the decision trees on official software HM16.2 and test the algorithm on the testing set. Experiments indicate that the fast decision algorithm improve the coding performance more efficiently than some existing algorithms.

Keywords: HEVC Inter prediction Data mining Decision trees

Published: 2018-10-12
Appears in: SpringerLink

: http://dx.doi.org/10.1007/978-3-030-00557-3_10