PROJECT TITLE :

Spatio-temporal mean curvature based image sequence restoration

ABSTRACT:

In this study, the authors propose a restoration algorithm for blurred and noisy continuous image sequences. The proposed approach treats a picture sequence as a space-time volume and employs a spatio-temporal mean curvature regularisation which is a novel regularisation proposed within the study to enhance the smoothness of the answer. An augmented Lagrangian technique with splitting techniques is used to handle the problem, iteratively finding solutions to the subproblems. Experiments show that the proposed approach will manufacture higher quality results and additional natural pictures comparing with other space-time volume based strategies on image sequence denoising and deblurring issues.