Real-Time Low Power Software HEVC Decoder on Embedded GPP: A Side-by-Side Comparison

Abstract : Video on mobile devices is a must-have feature with the prominence of new video-centric services and applications such as streaming or conferencing. By improving the compression performance by a factor of two with a similar image quality, the new video standard High Efficiency Video Coding (HEVC) is an appealing technology for service providers. Besides, with the recent progress of Systems-on-Chip (SoC) for mobile devices, software video decoders are now a reality. They offer a high flexibility of implementation as well as a means to rapidly deploy HEVC on existing devices without waiting for compatible hardware solutions. In this context, a challenge is to design a power efficient system that fits the compelling demand for long battery life on mobile devices. In this demo, a subjective evaluation of a low power HEVC decoder is presented on a big.LITTLE ARM-based embedded development platforms. It shows the performance of a standard compliant decoding system versus a low power decoding system in a side-by-side format on two 9-inch displays. Power savings close to 7% have been demonstrated on the whole energy consumption of a mobile device by using low power decoding.
Type de document :
Communication dans un congrès
Design and Architectures for Signal and Image Processing (DASIP), Sep 2015, Cracow, Poland. 2015
Liste complète des métadonnées

Littérature citée [9 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01205912
Contributeur : Erwan Raffin <>
Soumis le : lundi 28 septembre 2015 - 10:43:46
Dernière modification le : jeudi 7 février 2019 - 17:50:22
Document(s) archivé(s) le : mardi 29 décembre 2015 - 10:28:37

Fichier

raffin_demonight_dasip2015.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01205912, version 1

Citation

Erwan Raffin, Erwan Nogues, Morgan Lacour, Maxime Pelcat, Daniel Menard, et al.. Real-Time Low Power Software HEVC Decoder on Embedded GPP: A Side-by-Side Comparison. Design and Architectures for Signal and Image Processing (DASIP), Sep 2015, Cracow, Poland. 2015. 〈hal-01205912〉

Partager

Métriques

Consultations de la notice

512

Téléchargements de fichiers

269