A framelet-based algorithm for video enhancement

Abstract

Video clips are made up of many still frames. Most of the times, the frames are small perturbations of their neighboring frames. Recently, we proposed a framelet-based algorithm to enhance the resolution of any frames in a video clip by solving it as a super-resolution image reconstruction problem. In this paper, we extend the algorithm to video enhancement, where we compose a high-resolution video from a low-resolution one. An experimental result of our algorithm on a real video clip is given to illustrate the performance. © 2011 Springer Science+Business Media B.V.