Slimmable Video Codec. (arXiv:2205.06754v1 [eess.IV])

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 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), …

arxiv codec video