PT Unknown
AU Zhaocheng Liu
   Luis Herranz
   Fei Yang
   Saiping Zhang
   Shuai Wan
   Marta Mrak
   Marc Gorriz
TI Slimmable Video Codec
BT CVPR 2022 Workshop and Challenge on Learned Image Compression (CLIC 2022, 5th Edition)
PY 2022
BP 1742
EP 1746
DI 10.1109/CVPRW56347.2022.00183
AB Neural video compression has emerged as a novel paradigm combining trainable multilayer neural net-works 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 addition, models are usually optimized for a single RD tradeoff. Recent slimmable image codecs can dynamically adjust their model capacity to gracefully reduce the memory and computation requirements, without harming RD performance. In this paper we propose a slimmable video codec (SlimVC), by integrating a slimmable temporal entropy model in a slimmable autoencoder. Despite a significantly more complex architecture, we show that slimming remains a powerful mechanism to control rate, memory footprint, computational cost and latency, all being important requirements for practical video compression.
ER