iccv iccv2013 iccv2013-293 iccv2013-293-reference knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Tomer Michaeli, Michal Irani
Abstract: Super resolution (SR) algorithms typically assume that the blur kernel is known (either the Point Spread Function ‘PSF’ of the camera, or some default low-pass filter, e.g. a Gaussian). However, the performance of SR methods significantly deteriorates when the assumed blur kernel deviates from the true one. We propose a general framework for “blind” super resolution. In particular, we show that: (i) Unlike the common belief, the PSF of the camera is the wrong blur kernel to use in SR algorithms. (ii) We show how the correct SR blur kernel can be recovered directly from the low-resolution image. This is done by exploiting the inherent recurrence property of small natural image patches (either internally within the same image, or externally in a collection of other natural images). In particular, we show that recurrence of small patches across scales of the low-res image (which forms the basis for single-image SR), can also be used for estimating the optimal blur kernel. This leads to significant improvement in SR results.
[1] I. Begin and F. R. Ferrie. Blind super-resolution using a learning-based approach. In ICPR, 2004.
[2] H. Chang, D.-Y. Yeung, and Y. Xiong. Super-resolution through neighbor embedding. In CVPR, 2004.
[3] Y. C. Eldar and T. Michaeli. Beyond bandlimited sampling. IEEE Signal Process. Mag., 26(3):48–68, 2009.
[4] R. Fattal. Image upsampling via imposed edge statistics. ACM TOG, 26(3):95, 2007.
[5] R. Fergus, B. Singh, A. Hertzmann, S. T. Roweis, and W. T.
[6]
[7]
[8]
[9]
[10] Freeman. Removing camera shake from a single photograph. ACM TOG, 25(3):787–794, 2006. G. Freedman and R. Fattal. Image and video upscaling from local self-examples. ACM TOG, 30(2): 12, 2011. W. T. Freeman, T. R. Jones, and E. C. Pasztor. Examplebased super-resolution. IEEE Computer Graphics and Applications, 22(2):56–65, 2002. D. Glasner, S. Bagon, and M. Irani. Super-resolution from a single image. In ICCV, 2009. S. Harmeling, S. Sra, M. Hirsch, and B. Scholkopf. Multiframe blind deconvolution, super-resolution, and saturation correction via incremental EM. In ICIP, 2010. Y. He, K. H. Yap, L. Chen, and L. P. Chau. A soft MAP framework for blind super-resolution image reconstruction. Image and Vision Computing, 27(4):364 373, 2009. M. Irani and S. Peleg. Improving resolution by image registration. CVGIP, 53(3):231–239, 1991. N. Joshi, R. Szeliski, and D. J. Kriegman. PSF estimation using sharp edge prediction. In CVPR, 2008. 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. S. Mallat and G. Yu. Super-resolution with sparse mixing estimators. Trans. Signal Proc., 19(1 1):2889–2900, 2010. –
[11]
[12]
[13]
[14]
[15] 9955 11 ×× Figure 9: SR on images taken by an iPhone camera. Note the the bicycles, the tree and the patterns on the walls. Recovered kernels (marked in red at the bottom-left of the images) are 11 1 1(left image) and 13 13 (right images).
[16] P. Milanfar, editor. Super-resolution imaging. CRC Press, 2010.
[17] J. Qiao, J. Liu, and C. Zhao. A novel SVM-based blind super-resolution algorithm. In International Joint Conference on Neural Networks, 2006.
[18] F. Sroubek, G. Cristobal, and J. Flusser. Simultaneous superresolution and blind deconvolution. In Journal of Physics: Conference Series, volume 124, 2008.
[19] Q. Wang, X. Tang, and H. Shum. Patch based blind image super resolution. In ICCV, 2005.
[20] J. Yang, J. Wright, T. Huang, and Y. Ma. Image superresolution as sparse representation of raw image patches. In CVPR, 2008.
[21] R. Zeyde, M. Elad, and M. Protter. On single image scale-up using sparse-representations. In Curves and Surfaces, 2012.
[22] M. Zontak and M. Irani. Internal statistics of a single natural image. In CVPR, 201 1. 995522