FRANÇAIS
A showcase of ÉTS researchers’ publications and other contributions
SEARCH

Highly parallel HEVC motion estimation based on multiple temporal predictors and nested diamond search

Downloads

Downloads per month over past year

Hojati, Esmaeil, Franche, Jean-François, Coulombe, Stéphane et Vazquez, Carlos. 2018. « Highly parallel HEVC motion estimation based on multiple temporal predictors and nested diamond search ». In 2017 IEEE International Conference on Image Processing (ICIP) (Beijing, China, Sept. 17-20, 2017), pp. 2746-2750.
Compte des citations dans Scopus : 1.

[img]
Preview
PDF
Coulombe S 2018 16805.pdf - Accepted Version
Use licence: All rights reserved to copyright holder.

Download (655kB) | Preview

Abstract

Rate-constrained motion estimation (RCME) is the most computationally intensive task of H.265/HEVC encoding. Massively parallel architectures, such as graphics processing units (GPUs), used in combination with a multi-core central processing unit (CPU), provide a promising computing platform to achieve fast encoding. However, the dependencies in deriving motion vector predictors (MVPs) prevent the parallelization of prediction units (PUs) processing at a frame level. Moreover, the conditional execution structure of typical fast search algorithms is not suitable for GPUs designed for data-intensive parallel problems. In this paper, we propose a novel highly parallel RCME method based on multiple temporal motion vector (MV) predictors and a new fast nested diamond search (NDS) algorithm well-suited for a GPU. The proposed framework provides fine-grained encoding parallelism. Experimental results show that our approach provides reduced GPU load with better BD-Rate compared to prior full search parallel methods based on a single MV predictor.

Item Type: Conference proceeding
Professor:
Professor
Coulombe, Stéphane
Vázquez, Carlos
Affiliation: Génie logiciel et des technologies de l'information, Génie logiciel et des technologies de l'information
Date Deposited: 04 Jul 2018 12:56
Last Modified: 17 Jan 2020 21:06
URI: https://espace2.etsmtl.ca/id/eprint/16805

Actions (login required)

View Item View Item