Haichao Zhang David Wipf Yanning Zhang
Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2013
This paper presents a robust algorithm for estimating a single latent sharp image given either a single or multiple blurry and/or noisy observations. The underlying multi-image blind deconvolution problem is solved by linking all of the observations together via a Bayesian-inspired penalty function which couples the unknown latent image, blur kernels, and noise levels together in a unique way. This coupled penalty function enjoys a number of desirable properties, including a mechanism whereby the relative-concavity or shape is adapted as a function of the intrinsic quality of each blurry observation. In this way, higher quality observations may automatically contribute more to the final estimate than heavily degraded ones. The resulting algorithm, which requires no essential tuning parameters, can recover a high quality image from a set of observations containing potentially both blurry and noisy examples, without knowing a priori the degradation type of each observation. Experimental results on both synthetic and real-world test images clearly demonstrate the efficacy of the proposed method.
sparse blind deblurring, sparse blind deconvolution, coupled adaptive sparse prior, sparse estimation, non-convex optimization
Coupled Adaptive Sparse Penalty
Coupled: The penalty function is coupled over the latent sharp image, blur kernel and noise level, which is a crucial difference to the conventional separate penalties in the MAP approach.
Adaptive: Because of the coupling, the shape (sparsity prompting ability) of the penalty function is adjusted adaptively according to the estimated noise level and blur kernel, inducing an intrinsic mechanism of 'coarse-to-fine' estimation.
(click to see the animation)
Results compared with state-of-the-art methods
A. Dual Motion Deblurring: restoration with two blurry images
B. Dual Exposure Deblurring: restoration with blurry/noisy image pair
Related Publication and Software
Haichao Zhang, David Wipf and Yanning Zhang, Multi-Image Blind Deblurring Using a Coupled Adaptive Sparse Prior, CVPR 2013
David Wipf and Haichao Zhang, Revisiting Bayesian Blind Deconvolution, MSRA Tech. Report, 2013
H.Zhang, J. Yang, Y. Zhang, N Nasrabadi, T. Huang, Close the Loop: Joint Blind Image Restoration and Recognition with Sparse Representation Prior, ICCV 2011
 A. Rav-Acha and S. Peleg. Two motion blurred images are better than one. Pattern Recognition Letters, 26:311–317, 2005.
 J.-F. Cai, H. Ji, C. Liu, and Z. Shen. Blind motion deblurring using multiple images. J. Comput. Physics, 228(14):5057–5071, 2009.
 J. Chen, L. Yuan, C.-K. Tang, and L. Quan. Robust dual motion deblurring. In CVPR, 2008.
 F.Sroubek and P. Milanfar. Robust multichannel blind deconvolution via fast alternating minimization. IEEE Trans. on Image Processing, 21(4):1687–1700, 2012.
 O. Whyte, J. Sivic, A. Zisserman, and J. Ponce. Non-uniform deblurring for shaken images. International Journal of Computer Vision, 98(2):168–186, 2012.
 L. Yuan, J. Sun, and H.-Y. Shum. Image deblurring with blurred/noisy image pairs. In SIGGRAPH, 2007.
 X. Zhu, F. Sroubek, and P. Milanfar. Deconvolving PSFs for a better motion deblurring using multiple images. In ECCV, 2012.
 S. Cho and S. Lee. Fast motion deblurring. In SIGGRAPH ASIA, 2009.
 R. Fergus, B. Singh, A. Hertzmann, S. T. Roweis, and W. T. Freeman. Removing camera shake from a single photograph. In SIGGRAPH, 2006.
 A. Levin, Y. Weiss, F. Durand, and W. T. Freeman. Understanding and evaluating blind deconvolution algorithms. In CVPR, 2009.
 A. Levin, Y. Weiss, F. Durand, and W. T. Freeman. Efficient marginal likelihood optimization in blind deconvolution. In CVPR, 2011
 J. A. Palmer, D. P. Wipf, K. Kreutz-Delgado, and B. D. Rao. Variational EM algorithms for non-Gaussian latent variable models. In NIPS, 2006.