Awesome
Handling Kernel Uncertainty with CNNs
This repo provides results on benchmark datasets of our CVPR 2018 paper (Non-blind Deblurring: Handling Kernel Uncertainty with CNNs).
"Non-blind Deblurring: Handling Kernel Uncertainty with CNNs", Subeesh Vasu, Venkatesh Maligireddy and A. N. Rajagopalan, IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018), Salt Lake City, USA, June 2018. main paper, supple, poster
BibTeX
@inproceedings{vasu2018non,
author = {Vasu, Subeesh and Reddy Maligireddy, Venkatesh and Rajagopalan, AN},
title = {Non-Blind Deblurring: Handling Kernel Uncertainty With CNNs},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
pages={3272--3281},
year = {2018}}
Results
<p align="center"> <img src="images/nbd_syntheg_levin.png"> </p> <p align="center"> <img src="images/nbd_syntheg_lai.png"> </p> <p align="center"> <img src="images/nbd_realeg_lai.png"> </p>Visual comparison for NBD from inaccurate kernel estimates on Levin CVPR 2011 and Lai CVPR 2016 datasets. Levin TOG 2007, Krishnan NIPS 2009, Zoran ICCV 2011, Danielyan TIP 2012, Ji TIP 2012, Kheradmand TIP 2014, and Schmidt CVPR 2014 are existing works on NBD.
Results on public benchmark datasets
References
[Levin TOG 2007] A. Levin, R. Fergus, F. Durand, and W. T. Freeman.: Image and depth from a conventional camera with a coded aperture. TOG 2007
[Krishnan NIPS 2009] D. Krishnan and R. Fergus. Fast image deconvolution using hyper-laplacian priors. NIPS 2009.
[Zoran ICCV 2011] D. Zoran and Y. Weiss.: From learning models of natural image patches to whole image restoration, ICCV 2011
[Levin CVPR 2011] A. Levin, Y. Weiss, F. Durand, and W. T. Freeman.: Efficient marginal likelihood optimization in blind deconvolution. CVPR 2011
[Danielyan TIP 2012] A. Danielyan, V. Katkovnik, and K. Egiazarian.: Bm3d frames and variational image deblurring, TIP 2012
[Ji TIP 2012] H. Ji and K. Wang.: Robust image deblurring with an inaccurate blur kernel, TIP 2012
[Sun ICCP 2013] L. Sun, S. Cho, J. Wang, and J. Hays.: Edge-based blur kernel estimation using patch priors. ICCP 2013
[Kheradmand TIP 2014] A. Kheradmand and P. Milanfar.: A general framework for regularized, similarity-based image restoration. TIP 2014
[Schmidt CVPR 2014] U. Schmidt and S. Roth.: Shrinkage fields for effective image restoration. CVPR 2014
[Lai CVPR 2016] W.-S. Lai, J.-B. Huang, Z. Hu, N. Ahuja, and M.-H. Yang.: A comparative study for single image blind deblurring. CVPR 2016