Abstract
Field of Expert (FoE) [1], which is one of the most popular probabilistic models of natural image prior, has been successfully applied to super resolution. Piecewise smoothness imposed on natural images is, however, a relatively limited model for texture image. In the field of deep learning, various approaches for texture modeling using the Restricted Boltzmann Machine (RBM) achieves or surpasses the state-of-the-art on many tasks such as texture synthesis and inpainting. In this paper, we apply the convolutional RBM (cRBM) to learning a texture prior. The maximum a posteriori (MAP) framework is proposed to utilize the probabilistic texture model well. The experiment is done on the Brodatz Dataset, and our experimental results are shown to be comparable to those using FoE and other super resolution approaches.
Chapter PDF
Similar content being viewed by others
References
Schmidt, U., Gao, Q., Roth, S.: A Generative perspective on MRFs in low-level vision. In: Proc. IEEE Conference on Computer Vision and Pattern Recognition, pp. 1751–1758 (2010)
Zhang, H., Zhang, Y., Li, H., Huang, T.: Generative Bayesian Image Super Resolution with Natural Image Prior. IEEE Transactions on Image Processing, 4054–4067 (2012)
Heess, N. Williams, C., Hinton, G.: Learning generative texture models with extended fields-of-experts. In: Proc. the British Machine Vision Conference (2009)
Luo, H. Carrier, P.L., Couville, A., Bengio, Y.: Texture modeling with convolutional spike-and-slab RBMs and deep extensions. In: Proc. the International Conference on Artificial Intelligence and Statistics (AISTAT), pp. 415–423 (2013)
Kivinen, J., Williams, C.: Multiple texture Boltzmann machines. In: Proc. the International Conference on Artificial Intelligence and Statistics (AISTAT), pp. 4054–4067 (2012)
Gao, Qi, Roth, Stefan: Texture synthesis: from convolutional RBMs to efficient deterministic algorithms. In: Fränti, Pasi, Brown, Gavin, Loog, Marco, Escolano, Francisco, Pelillo, Marcello (eds.) S+SSPR 2014. LNCS, vol. 8621, pp. 434–443. Springer, Heidelberg (2014)
Ranzato, M., Mnih, V., Susskind, J., Hinton, G.: Modeling Natural Images Using Gated MRFs. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2206–2226 (2013)
Lee, H., Grosse, R., Ranganath, R., Ng, A.Y.: Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations. In: Proc. International Conference on Machine Learning, pp. 609–616 (2009)
Nair, V., Hinton, G.: Implicit mixtures of restricted Boltzmann machines. In: Proc. Advances in Neural Information Processing System (2009)
Welling, M., Rosen-Zvi, M., Hinton, G.: Exponential family harmoniums with an application to information retrieval. In: Proc. Advances in Neural Information Processing System (2008)
Takeda, H., Farisu, S., Milanfar, P.: Kernel Regression for Image Processing and Reconstruction. IEEE Transactions on Image Processing 12(2), 349–336 (2007)
Glasner, D., Bagon, S., Irani, M.: Super-resoluton from a single image. In: Proc. IEEE Conference on Computer Vision and Pattern Recognition (2009)
Freeman, G., Fattal, R.: Image and Video Upscaling from Local Self-Examples. ACM Trans. Graph. 28(3), 1–10 (2009)
Yang, J., Wang, Z., Lin, Z., Huang, T.: Coupled dictionary training for image super resolution. IEEE Transactions on Image Processing 21(8), 3467–3478 (2012)
Kim, K.I., Kwon, Y.: Single-image Super-resolution Using Sparse Regression and Natural Image Prior. IEEE Transaction on Pattern Analysis and Machine Intelligence (2008)
Ngiam, J., Chen, Z., Koh, P.W., Ng, A.Y.: Learning deep energy models. In: Proc. International Conference on Machine Learning (2011)
Wang, Z., Bovik, A.C.: Mean squared error: Love it or leave it? A New look at Signal Fidelity Measures. IEEE Signal Process. Mag. 26(1), 98–117 (2009)
Tieleman, T.: Training restricted boltzmann machines using approximations to the likelihood gradient. In: Proc. International Conference on Machine Learning (2008)
Neal, R.M.: MCMC using Hamiltonian dynamics (2012). arXiv:1206.1901v1
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Kang, C., Hong, M., Yoo, S.I. (2015). Learning Texture Image Prior for Super Resolution Using Restricted Boltzmann Machine. In: Murino, V., Puppo, E. (eds) Image Analysis and Processing — ICIAP 2015. ICIAP 2015. Lecture Notes in Computer Science(), vol 9279. Springer, Cham. https://doi.org/10.1007/978-3-319-23231-7_20
Download citation
DOI: https://doi.org/10.1007/978-3-319-23231-7_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-23230-0
Online ISBN: 978-3-319-23231-7
eBook Packages: Computer ScienceComputer Science (R0)