Adaptive regularization-based space–time super-resolution reconstruction

• We propose an adaptive space–time regularization method for multiple video sequence super-resolution reconstruction.
• An alternating minimization algorithm is employed to solve the proposed method.
• The proposed method is efficient for different video sequences.

In this paper, we propose an adaptive space–time regularization method for multiple video sequence super-resolution reconstruction. In the proposed method, the regularization item is a functional with the variable exponent, 1≤p≤21≤p≤2. The local exponent p is selected according to the space–time gradient of each voxel of the video sequence, and the local regularization parameter is selected according to the space–time activity of each voxel of the video sequence. An alternating minimization algorithm is employed to solve the proposed method. Experiment results have verified the effectiveness of the proposed method and demonstrated its superiority to other regularization methods.