The student-t mixture as a natural image patch prior with application to image compression

Recent results have shown that Gaussian mixture models (GMMs) are remarkably good at density modeling of natural image patches, especially given their simplicity. In terms of log likelihood on real-valued data they are comparable with the best performing techniques published, easily outperforming more advanced ones, such as deep belief networks. They can be applied to various image processing tasks, such as image denoising, deblurring and inpainting, where they improve on other generic prior methods, such as sparse coding and field of experts. Based on this we propose the use of another, even richer mixture model based image prior: the Student-t mixture model (STM). We demonstrate that it convincingly surpasses GMMs in terms of log likelihood, achieving performance competitive with the state of the art in image patch modeling. We apply both the GMM and STM to the task of lossy and lossless image compression, and propose efficient coding schemes that can easily be extended to other unsupervised machine learning models. Finally, we show that the suggested techniques outperform JPEG, with results comparable to or better than JPEG 2000.

[1]  D. Anderson,et al.  Algorithms for minimization without derivatives , 1974 .

[2]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[3]  A. F. Smith,et al.  Statistical analysis of finite mixture distributions , 1986 .

[4]  A. F. Smith,et al.  Statistical analysis of finite mixture distributions , 1986 .

[5]  Gregory K. Wallace,et al.  The JPEG still picture compression standard , 1991, CACM.

[6]  David J. Field,et al.  Sparse coding with an overcomplete basis set: A strategy employed by V1? , 1997, Vision Research.

[7]  Geoffrey J. McLachlan,et al.  Robust mixture modelling using the t distribution , 2000, Stat. Comput..

[8]  Jan Skoglund,et al.  Vector quantization based on Gaussian mixture models , 2000, IEEE Trans. Speech Audio Process..

[9]  Touradj Ebrahimi,et al.  The JPEG 2000 still image compression standard , 2001, IEEE Signal Process. Mag..

[10]  Jitendra Malik,et al.  A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[11]  Vivek K. Goyal,et al.  Theoretical foundations of transform coding , 2001, IEEE Signal Process. Mag..

[12]  Geoffrey E. Hinton,et al.  Learning Sparse Topographic Representations with Products of Student-t Distributions , 2002, NIPS.

[13]  Geoffrey E. Hinton.  Training Products of Experts by Minimizing Contrastive Divergence , 2002, Neural Computation.

[14]  Gerald Schaefer,et al.  UCID: an uncompressed color image database , 2003, IS&T/SPIE Electronic Imaging.

[15]  Samuel Kotz,et al.  Multivariate T-Distributions and Their Applications , 2004 .

[16]  Michael J. Black,et al.  Fields of Experts: a framework for learning image priors , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[17]  Robert M. Gray,et al.  Lloyd clustering of Gauss mixture models for image compression and classification , 2005, Signal Process. Image Commun..

[18]  Kun Huang,et al.  A multiscale hybrid linear model for lossy image representation , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[19]  Michael Elad,et al.  Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries , 2006, IEEE Transactions on Image Processing.

[20]  William T. Freeman,et al.  What makes a good model of natural images? , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[21]  Michael Elad,et al.  Compression of facial images using the K-SVD algorithm , 2008, J. Vis. Commun. Image Represent..

[22]  Guillermo Sapiro,et al.  Non-local sparse models for image restoration , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[23]  Yoshua Bengio,et al.  Why Does Unsupervised Pre-training Help Deep Learning? , 2010, AISTATS.

[24]  Guillermo Sapiro,et al.  Sparse Representation for Computer Vision and Pattern Recognition , 2010, Proceedings of the IEEE.

[25]  Christine Guillemot,et al.  Image Compression Using Sparse Representations and the Iteration-Tuned and Aligned Dictionary , 2011, IEEE Journal of Selected Topics in Signal Processing.

[26]  Yair Weiss,et al.  From learning models of natural image patches to whole image restoration , 2011, 2011 International Conference on Computer Vision.

[27]  Matthias Bethge,et al.  In All Likelihood, Deep Belief Is Not Enough , 2010, J. Mach. Learn. Res..

[28]  William A. Pearlman,et al.  Digital Signal Compression: Principles and Practice , 2011 .

[29]  Yair Weiss,et al.  "Natural Images, Gaussian Mixtures and Dead Leaves" , 2012, NIPS.

[30]  M. Bethge,et al.  Mixtures of Conditional Gaussian Scale Mixtures Applied to Multiscale Image Representations , 2011, PloS one.

[31]  Stéphane Mallat,et al.  Solving Inverse Problems With Piecewise Linear Estimators: From Gaussian Mixture Models to Structured Sparsity , 2010, IEEE Transactions on Image Processing.

[32]  I. Horev,et al.  Adaptive image compression using sparse dictionaries , 2012, 2012 19th International Conference on Systems, Signals and Image Processing (IWSSIP).

[33]  Geoffrey E. Hinton,et al.  Deep Mixtures of Factor Analysers , 2012, ICML.

[34]  Hugo Larochelle,et al.  RNADE: The real-valued neural autoregressive density-estimator , 2013, NIPS.

[35]  Geoffrey E. Hinton,et al.  Tensor Analyzers , 2013, ICML.

[36]  Benjamin Schrauwen,et al.  Learning a piecewise linear transform coding scheme for images , 2013, International Conference on Graphic and Image Processing.

[37]  Hugo Larochelle,et al.  A Deep and Tractable Density Estimator , 2013, ICML.