Spatio-Temporal Fusion Network for Video Super-Resolution

Huabin Li,Pingjian Zhang
DOI: https://doi.org/10.1109/IJCNN52387.2021.9533610
2021-01-01
Abstract:The video super-resolution (VSR) is a traditional computer vision task, and in recent years, the methods based on deep learning have achieved promising performance. Most of the previous VSR methods utilize optical flow to perform alignment. However, in the case of occlusions and large motions, it is difficult to obtain accurate optical flow, which will affect the final recovery results. In additio...
What problem does this paper attempt to address?