1 / 23

Image Super-Resolution as Sparse Representation of Raw Image Patches

Image Super-Resolution as Sparse Representation of Raw Image Patches. Jianchao Yang, John Wright, Thomas Huang, Yi Ma CVPR 2008. Outline. Introduction Super-resolution from Sparsity Local Model from Sparse Representation Enforcing Global Reconstruction Constraint

felix
Download Presentation

Image Super-Resolution as Sparse Representation of Raw Image Patches

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Image Super-Resolution as Sparse Representation of Raw Image Patches Jianchao Yang, John Wright, Thomas Huang, Yi Ma CVPR 2008

  2. Outline • Introduction • Super-resolution from Sparsity • Local Model from Sparse Representation • Enforcing Global Reconstruction Constraint • Global Optimization Interpretation • Dictionary Preparation • Experiments • Discussion

  3. Introduction • To generate a super-resolution (SR) image requires multiple low-resolution images of the same scene, typically aligned with sub-pixel accuracy • MAP (maximum a-posteriori) • Markov Random Field (MRF) solved by belief propagation • Bilateral Total Variation • The SR task is cast as the inverse problem of recovering the original high-resolution image by fusing the low-resolution images

  4. Introduction • Let be an overcomplete dictionary of K prototype signal-atoms • Suppose a signal can be represented as a sparse linear combination of these atoms • The signal vector can be written as , whereis a vector with very few (<<K) nonzero entries • In practice, we might observe only a small set of measurements of : where with

  5. Introduction • In the super-resolution context, is a high-resolution image (patch), while is its low-resolution version (or features extracted from it) • If the dictionary is overcomplete, the equation is underdetermined for the unknown coefficients α • Same as • In our setting, using two coupled dictionaries • for high-resolution patches • for low-resolution patches

  6. Super-resolution from Sparsity • The single-image super-resolution problem • Given a low-resolution image Y , recover a higher-resolution image X of the same scene • Reconstruction constraint • The observed low-resolution image Y is a blurred and downsampled version of the solution X • :H represents a blurring filterD is the downsampling operator

  7. Super-resolution from Sparsity • Sparse representation prior • The patches x of the high-resolution image X can be represented as a sparse linear combination in a dictionary of high-resolution patches sampled from training images

  8. Local Model from Sparse Representation • For this local model, we have two dictionaries and • is composed of high-resolution patches • is composed of corresponding low-resolution patches • Subtract the mean pixel value for each patch :the dictionary represents image textures rather than absolute intensities • For each input low-resolution patch y, we find a sparse representation with respect to • Reconstruct x by the corresponding high-resolution patch

  9. Local Model from Sparse Representation • The problem of finding the sparsest representation of y can be formulated as : • Fis a (linear) feature extraction operator • (4) is NP-hard • Can be efficiently recovered by instead minimizing the -norm • Lagrange multipliers offer an equivalent formulation

  10. Local Model from Sparse Representation • Pextracts the region of overlap between current target patch and previously reconstructed high-resolution image • w contains the values of the previously reconstructed high-resolution image on the overlap

  11. Enforcing Global Reconstruction Constraint • (5) and (7) do not demand exact equality between the low-resolution patch y and its reconstruction • Because of noise • The solution to this optimization problem can be efficiently computed using the back-projection method • The update equation for this iterative method is

  12. Enforcing Global Reconstruction Constraint

  13. Global Optimization Interpretation • Applied image compression, denoising, and restoration [17] • This leads to a large optimization problem [17] J. Mairal, G. Sapiro, and M. Elad. Learning multiscale sparse representations for image and video restoration. SIAM Multiscale Modeling and Simulation, 2008.

  14. Dictionary Preparation • Random Raw Patches from Training Images • we generate dictionaries by simply randomly sampling raw patches from training images of similar statistical nature • For our experiments, we prepared two dictionaries • one sampled from flowers • one sampled from animal images • For each highresolution training image X, we generate the corresponding low-resolution image Y by blurring and downsampling • For each category of images, we sample only about 100,000 patches from about 30 training images to form each dictionary

  15. Dictionary Preparation • Derivative Features • use a feature transformation F to ensure that the computed coefficients fit the most relevant part of the low-resolution signal

  16. Experimental settings • In our experiments, we will mostly magnify the input image by a factor of 3 • In the low-resolution images, we use 3 × 3 low-resolution patches, with overlap of 1 pixel between adjacent patches, corresponding to 9 × 9 patches with overlap of 3 pixels for the high-resolution patches • For color images, we apply our algorithm to the illuminance component • λ = 50 × dim (patch feature)

  17. Experimental results [5] H. Chang, D.-Y. Yeung, and Y. Xiong. Super-resolution through neighbor embedding. CVPR, 2004.

  18. Experimental results [6] S. Dai, M. Han, W. Xu, Y. Wu, and Y. Gong. Soft edge smoothness prior for alpha channel super resolution. Proc. ICCV, 2007.

  19. Experimental results

  20. Experimental results

  21. Experimental results [12] W. T. Freeman, E. C. Pasztor, and O. T. Carmichael. Learning low-level vision. IJCV, 2000.

  22. Experimental results

  23. Discussion • The experimental results demonstrate the effectiveness of sparsity as a prior for patch-based super-resolution • One of the most important questions for future investigation is to determine the number of raw sample patches required to generate a dictionary • Tighter connections to the theory of compressed sensing may also yield conditions on the appropriate patch size or feature dimension

More Related