EE SEminar: Blind Image Deblurring Via Residual Blur Elimination

~~Speaker: Rana Hanocka, 
M.Sc. student under the supervision of Prof. Nahum Kiryati

Wednesday, December 16th, 2015 at 15:30
Room 011, Kitot Bldg., Faculty of Engineering

Blind Image Deblurring Via Residual Blur Elimination

Abstract

We present a novel progressive framework for blind image restoration. Common blind restoration schemes first estimate the blur kernel, then employ non-blind deblurring. However, despite recent progress, the accuracy of PSF estimation is limited. Furthermore, the outcome of non-blind deblurring is highly sensitive to errors in the assumed PSF. Therefore, high quality blind deblurring has remained a major challenge. We suggest an iterative progressive restoration scheme, in which the imperfectly deblurred output of the current iteration is fed back as input to the next iteration. The kernel representing the residual blur is then estimated, and used to drive the non-blind restoration component, leading to finer deblurring. Our framework is extremely modular. Our proposed method is able to adapt to various kernel estimation and non-blind deblurring modules, including state-of-the-art art regularizers for the image and the PSF. A particularly interesting combination is the Mumford & Shah piecewise-smooth image model and the sparse PSF prior. Previous works that used Mumford & Shah image regularization were either limited to non-blind deblurring or semi-blind deblurring assuming a parametric kernel known up to an unknown parameter. Experimental results demonstrate rapid convergence, and excellent performance on a wide variety of blurred images.

16 בדצמבר 2015, 15:30 
חדר 011, בניין מעבדות חשמל 
אוניברסיטת תל אביב עושה כל מאמץ לכבד זכויות יוצרים. אם בבעלותך זכויות יוצרים בתכנים שנמצאים פה ו/או השימוש
שנעשה בתכנים אלה לדעתך מפר זכויות, נא לפנות בהקדם לכתובת שכאן >>