Efficient HEVC-to-VVC Transcoder Based On A Bayesian Classifier For The First Quadtree Depth Level.pdf (1.2 MB)
Download file

Efficient HEVC-to-VVC Transcoder Based On A Bayesian Classifier For The First Quadtree Depth Level

Download (1.2 MB)
conference contribution
posted on 25.03.2022, 16:10 authored by D Garcia-Lucas, G. Cebrián-Márquez, A. J. Díaz-Honrubia, Thanuja MallikarachchiThanuja Mallikarachchi, P. Cuenca
In the coming years, the Versatile Video Coding (VVC) standard will be launched to replace the current High Efficiency Video Coding (HEVC) standard, making it necessary to find efficient methods to convert existing multimedia content to the new format. However, transcoding is a complex pipeline composed of a decoding and an encoding process that involves long processing times. On the basis of the existing correlation between the block partitioning structures of both standards, this paper presents an HEVC-to-VVC transcoding scheme. The proposed method consists of a Naïve-Bayes classifier that assists the partitioning decision at the first level of quadtree by using features extracted from the 128×128 pixel blocks of the residual and reconstructed frames in HEVC. The experimental results using random access configuration show an average transcoding time reduction of 13.38% at the cost of a compression efficiency loss of 0.32% in terms of BD-rate.

History

Presented at

Conference paper published in proceedings of 2020 IEEE International Conference on Image Processing (ICIP)

Link to Conference Website

Published in

2020 IEEE International Conference on Image Processing (ICIP)

Publisher

IEEE

Version

AM (Accepted Manuscript)

Citation

D. García-Lucas, G. Cebrián-Márquez, A. J. Díaz-Honrubia, T. Mallikarachchi and P. Cuenca (2020) 'Efficient HEVC-to-VVC Transcoder Based On A Bayesian Classifier For The First Quadtree Depth Level,' 2020 IEEE International Conference on Image Processing (ICIP), Abu Dhabi, United Arab Emirates, 2020, pp. 628-632, doi: 10.1109/ICIP40778.2020.9190640.

Print ISSN

1522-4880

Electronic ISSN

2381-8549

ISBN

978-1-7281-6395-6

Cardiff Met Affiliation

  • Cardiff School of Technologies

Cardiff Met Authors

Thanuja Mallikarachchi

Copyright Holder

© The Publisher

Publisher Rights Statement

© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

Language

en