Space-Time Video Super-resolution with Neural Operator

arXiv:2404.06036v1 Announce Type: new
Abstract: This paper addresses the task of space-time video super-resolution (ST-VSR). Existing methods generally suffer from inaccurate motion estimation and motion compensation (MEMC) problems for large motions. Inspired by recent progress in physics-informed neural networks, we model the challenges of MEMC in ST-VSR as a mapping between two continuous function spaces. Specifically, our approach transforms independent low-resolution representations in the coarse-grained continuous function space into refined representations with enriched spatiotemporal details in the fine-grained continuous function …

abstract arxiv challenges compensation continuous cs.cv function mapping networks neural networks paper physics physics-informed progress resolution space type video