Characterizing and removing motion blur caused by camera shake or objectmotion remains an important task for image restoration . We propose a general,non-parametric model for dense non-uniform motion blur estimation . We estimate a set of adaptive basis kernels as well as the mixingcoefficient at pixel level, producing a per-pixel map of motion blur . Thisrich but efficient forward model of the degradation process allows theutilization of existing tools for solving inverse problems . We show that ourmethod overcomes the limitations of existing non-Uniform motion . estimationand that it contributes to bridging the gap between model-based and data-drivenapproaches for deblurring real photographs for real photographs . We hope to use this method to improve the accuracy of our image restoration

Author(s) : Guillermo Carbajal, Patricia Vitoria, Mauricio Delbracio, Pablo Musé, José Lezama

Links : PDF - Abstract

Code :
Coursera

Keywords : motion - blur - image - uniform - model -

Leave a Reply

Your email address will not be published. Required fields are marked *