Camera Intrinsic Blur Kernel Estimation: A Reliable Framework

Ali Mosleh, Paul Green, Emmanuel Onzon, Isabelle Begin, J.M. Pierre Langlois; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 4961-4968

Abstract


This paper presents a reliable non-blind method to measure intrinsic lens blur. We first introduce an accurate camera-scene alignment framework that avoids erroneous homography estimation and camera tone curve estimation. This alignment is used to generate a sharp correspondence of a target pattern captured by the camera. Second, we introduce a Point Spread Function (PSF) estimation approach where information about the frequency spectrum of the target image is taken into account. As a result of these steps and the ability to use multiple target images in this framework, we achieve a PSF estimation method robust against noise and suitable for mobile devices. Experimental results show that the proposed method results in PSFs with more than 10 dB higher accuracy in noisy conditions compared with the PSFs generated using state-of-the-art techniques.

Related Material


[pdf]
[bibtex]
@InProceedings{Mosleh_2015_CVPR,
author = {Mosleh, Ali and Green, Paul and Onzon, Emmanuel and Begin, Isabelle and Pierre Langlois, J.M.},
title = {Camera Intrinsic Blur Kernel Estimation: A Reliable Framework},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2015}
}