Video Super Resolution Based on Deep Learning: A Comprehensive Survey

• DOI: 10.48550/arxiv.2007.12928
• Publication Date: 2020-01-01
• Authors: Hongying Liu; Zhubo Ruan; Peng Zhao; Chao Dong; Fanhua Shang; Yuanyuan Liu; Linlin Yang; Radu Timofte
• Keywords: Leverage (statistics); Benchmark (surveying)

In recent years, deep learning has made great progress in many fields such as image recognition, natural language processing, speech recognition and video super-resolution. this survey, we comprehensively investigate 33 state-of-the-art super-resolution (VSR) methods based on learning. It is well known that the leverage of information within frames important for Thus propose a taxonomy classify into six sub-categories according to ways utilizing inter-frame information. Moreover, architectures implementation details all are depicted detail. Finally, summarize compare performance representative VSR method some benchmark datasets. We also discuss challenges, which need be further addressed by researchers community VSR. To best our knowledge, work first systematic review tasks, it expected make contribution development studies area potentially deepen understanding techniques