Nous vous rappelons que, afin de garantir l'accès de tous les inscrits aux salles de réunion, l'inscription aux réunions est gratuite mais obligatoire.
Inscriptions closes à cette réunion.
28 personnes membres du GdR ISIS, et 45 personnes non membres du GdR, sont inscrits à cette réunion.
Capacité de la salle : 100 personnes.
Les comités de standardisation ISO/MPEG et ITU/VCEG ont conjointement mis en place un groupe de travail Joint Video Experts Team (JVET) pour le développement d'une nouvelle norme de compression vidéo appelée VVC pour Versatile Video Coding. La norme VVC qui a été finalisée en Juillet 2020 permet une réduction de la bande passante de 50% par rapport à la norme précédente HEVC pour une qualité visuelle équivalente. Néanmoins, ce gain en débit-distorsion est obtenu au détriment d'une augmentation considérable de la complexité d'encodage comparé à HEVC.
L'objectif de cette journée est de présenter des travaux de recherche développés autour de cette nouvelle norme. Ces travaux concernent non seulement les outils de codage proposés dans le cadre des activités de standardisation mais aussi des techniques de réduction de la complexité d'encodage VVC, les solutions logicielles et matérielles de codage et de décodage temps réel ainsi que des algorithmes non normatifs permettant d'améliorer la qualité d'une vidéo VVC.
Benjamin Bross - HHI
Versatile Video Coding (VVC) is the most recent jointly developed video coding standard by ITU-T (H.266) and ISO/IEC (MPEG-I Part 3). It is based on the same block-based hybrid video coding design as its predecessors from MPEG-2 to H.265/HEVC. VVC is designed to be both efficient and versatile to address today's media needs. This includes approximately 50% bit-rate reduction over H.265/HEVC as well as versatility by efficient coding of a wide range of video content and applications. In this talk, video coding techniques are reviewed including tools representing an evolution of known techniques but also more revolutionary tools developed using machine learning.
T. Biatek, M. Abdoli, T. Guionnet, A. Nasrallah and M. Raulet - ATEME
8K-TV momentum has grown these past years, fostered by CE-display manufacturers and perspective of Tokyo?s Olympics broadcasting. However, a broad 8K-TV deployment is still uncertain. Although HEVC provides sufficient coding efficiency to enable DTH broadcasting, the transmission cost remains high and the HEVC licensing situation makes deployment complicated, especially for DTT. In that context, the emerging codecs VVC and EVC are both capable of addressing these issues by increasing coding efficiency without repeating HEVC licensing situation. In this paper, we demonstrate how VVC and EVC could be 8K-broadcast enablers in the upcoming years. Based on encoding constraints coming from DVB-T2/S2 and 5G-broadcast transmission scenarios, the relevance of both codecs is assessed based on encoding efficiency and complexity criterions. In addition, we highlight that early 8K-deployment is possible with these codecs since a reduced set of tools is capable of achieving minimal required efficiency. Finally, some preliminary results of ATEME industrial VVC encoding platform are provided to show that early 8K deployment is possible using the latest video coding standards.
Naty Sidaty, Wassim Hamidouche, Pierrick Philippe and Jérôme Fournier and Olivier Déforges - IRT b<>com and Univ Rennes, INSA Rennes, CNRS, IETR - UMR 6164
Video compression and content quality have become one of the most research topic in the recent years. Predominantly, trends obviously signpost that the video usage over the Internet is on the upsurge. Simultaneously, users? requirement for enlarged resolution and higher quality is rising. Consequently, a huge effort has been made for video coding technologies and quality monitoring. We present a subjective-based comparison as well as an objective measurement between the newest Versatile Video Coding (VVC) and the well-known High Efficiency Video Coding (HEVC) standards. Several videos of various content are selected as tested sequences. Both High Definition (HD) and Ultra High Definition (UHD) resolutions are used in this experiment. An extensive range of bit-rates from low to high bit-rates were selected. These sequences are encoded using both HEVC reference software (HM-16.2) and the latest reference software of VVC (VTM-5.0). Obtained results have shown that VVC outperforms consistently HEVC, for realistic bit rates and quality levels, in the range of 40% on the subjective scale. For the objective measurements, using PSNR, SSIM and VMAF as quality metrics, the quality enhancement of VVC over HEVC is ranging from 31% to 40%, depending on video content and spatial resolution.
Edouard François, Michel Kerdranvat, Rémi Jullian, Christophe Chevance, Philippe de Lagrange, Fabrice Urban, Tangi Poirier, Ya Chen - Interdigital
VVC est la plus récente norme de codage vidéo (VVC) développée conjointement par MPEG (ISO / IEC) et VCEG (ITU-T). VVC a été développé incrémentalement sur une base HEVC, avec l'introduction de nouveaux outils de codage sur l?ensemble des éléments de l?architecture du codec.
Cette présentation propose une analyse des gains en compression et de la complexité de ces différents outils. Elle se base sur une évaluation du gain de codage de chaque outil pour divers contenus. Une comparaison de performance de VVC par rapport à HEVC est aussi présentée.
Adam Wieckowski - HHI
The Versatile Video Coding (H.266/VVC) standard was finalized in June 2020, being jointly developed by ITU-T and ISO/IEC. The codec delivers around 50% subjective quality improvement at the same rate with regards to its predecessor HEVC/H.265. The coding gain comes at the expense of more complicated algorithms. During the development, an emphasis was put on keeping the decoding complexity low. The VVC complexity increase over HEVC, as measured by the reference software runtime, is around 2x for the decoder and 10x for the encoder. This talk will discuss the Fraunhofer Versatile Video Encoder&Decoder implementations, released just two months after standard finalization. For the decoder, an optimized an parallelized implementation is discussed, also touching on where the VVC complexity comes from. For the encoder, the optimized implementation has much less constraints. The talk will discuss on what approach we took to take most of what VVC has to offer at lowest possible complexity.
Date : 2020-10-20
Lieu : Visio-Conférence
Thèmes scientifiques :
C - Algorithme-architecture en traitement du signal et des images
D - Télécommunications : compression, protection, transmission
Inscriptions closes à cette réunion.
Accéder au compte-rendu de cette réunion.
(c) GdR IASIS - CNRS - 2024.