A Learnable EVC Intra Predictor Using Masked Convolutions - CentraleSupélec Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

A Learnable EVC Intra Predictor Using Masked Convolutions

Roberto Iacoviello
Alessandra Mosca
  • Fonction : Auteur

Résumé

The Enhanced Video Coding (EVC) workgroup of the Moving Picture, Audio and Data Coding by Artificial Intelligence (MPAI) organization aims at enhancing traditional video codecs by improving or replacing traditional encoding tools with AI-based counterparts. In this work, we explore enhancing MPEG Essential Video Coding (EVC) intra prediction with a learnable predictor: we recast the problem as a hole inpainting task that we tackle via masked convolutions. Our experiments in standard test conditions show BD-rate reductions in excess of 6% over the EVC baseline profile reference with some sequences in excess of 12%.
Fichier principal
Vignette du fichier
ICIAP_23_EVC_Intra (1).pdf (430.82 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04201123 , version 1 (09-09-2023)

Identifiants

Citer

Gabriele Spadaro, Roberto Iacoviello, Alessandra Mosca, Giuseppe Valenzise, Attilio Fiandrotti. A Learnable EVC Intra Predictor Using Masked Convolutions. International Conference on Image Analysis and Processing, Nov 2023, Udine, Italy. pp.537-549, ⟨10.1007/978-3-031-43148-7_45⟩. ⟨hal-04201123⟩
14 Consultations
36 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More