We address the problem of blind piecewise space-variant image deblurring where only part of the image is sharp, assuming a shallow depth of field which imposes significant defocus blur. Based on a special blur measure, we propose an automatic image recovery approach which segments the sharp and blurred sub-regions, estimates a non-parametric blur kernel and restores the sharp image via a variational space-variant method. Experimental results on real images demonstrate the effectiveness of the algorithm.

This presentation is part of Minisymposium “MS43 - Variational Image Segmentation: Methods and Applications”
organized by: Jack Spencer (University of Liverpool) .

Authors:

Leah Bar (Tel-Aviv University and maxq.ai )

Nir Sochen (School of Mathematics, Tel-Aviv University)

Nahum Kiryati (School of Electrical Engineering, Tel-Aviv University)

Keywords: