Slimmable Video Codec

Neural video compression has emerged as a novel paradigm combining trainable multilayer neural networks and machine learning, achieving competitive rate-distortion (RD) performances, but still remaining impractical due to heavy neural architectures, with large memory and computational demands. In ad...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2022-05
Hauptverfasser:	Liu, Zhaocheng, Herranz, Luis, Yang, Fei, Zhang, Saiping, Wan, Shuai, Mrak, Marta, Marc Górriz Blanch
Format:	Artikel
Sprache:	eng
Schlagworte:	Codec Computing costs Machine learning Multilayers Network latency Neural networks Video compression
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!