1 / 31

Vector Valued Image Regularization with PDE’s: A Common Framework for Different Applications

Vector Valued Image Regularization with PDE’s: A Common Framework for Different Applications. CVPR 2003 Best Student Paper Award. Definitions. Scalar images: Intensity images Vector valued images: RGB, HSV, YIQ… Regularization: Finding approximate solution for ill-posed problems.

Download Presentation

Vector Valued Image Regularization with PDE’s: A Common Framework for Different Applications

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. Vector Valued Image Regularization with PDE’s: A Common Framework for Different Applications CVPR 2003 Best Student Paper Award

  2. Definitions • Scalar images: Intensity images • Vector valued images: RGB, HSV, YIQ… • Regularization: Finding approximate solution for ill-posed problems. • Let G be 2x2 matrix

  3. Divergence and Curl vector function Divergence defines expansion or contraction per unit volume Vectors are obtained from image gradients I Curl of a vector field is defined by cross product between vectors

  4. Structure Tensor

  5. Structure Tensor One dimensional image (gray level)

  6. Structure Tensor Eigenvalue and eigenvectors of G (spectral elements) Can be computed using Matlab functions

  7. y derivative x derivative

  8. More Insight (hessian&tensor) • Hessian of image I • Laplacian of I • Tensor: Matrix of matrices. They abuse the notation: Symmetric semi-positive definite matrix

  9. yy derivative xx derivative

  10. yy derivative trace xx derivative

  11. Image Regularization • Functional minimization: Euler-Lagrange equations • Divergence expression: Diffusion of pixel values from high to low concentration • Oriented Laplacians: Image smoothing in eigenvector directions weighted by corresponding eigenvalue.

  12. Variational Problem • Define variational problem: Euler-Lagrange equation • Solution using classic iterative method:

  13. Variational Problem Horn&Schunck Example

  14. Defining Energy Functional • Functional: (minimization based) increasing function • Euler-Lagrange is given by (relation to divergence based methods)

  15. D Matrix is 2x2. Eigenvalues and eigenvectors of D are

  16. Old School Laplacian Approach Second order image derivatives in directions of eigenvectors of G at that point (structure tensor) (Edge preserving smoothing) Solution of this PDE can be given by:

  17. Laplacian Example (constant T) Constant T, such that direction (eigenvectors are same)

  18. Laplacian Example

  19. Laplacian Example (varying T) T is not constant and but independent of image content. There are similarities with bilateral filtering.

  20. Laplacian Example (varying T)

  21. Relation Between T and D Homework Due 19 January 2005 Show the following: (Page 3 of the paper):

  22. Relation Between T and D

  23. Relation Between T and D Let’s just solve for div(G). Rest can be solved similarly:

  24. Relation Between T and D Using trace property: Hessian Kronecker func.

  25. Rewriting the Formula super matrix notation

  26. Unified Expression

  27. Numerical Implementation • Conventional approach • Compute image Hessians and Gradients • Proposed method • Use local filtering by Gaussians (2nd page) Similar to bilateral filtering

  28. Numerical Implementation • Compute local convolution mask defining local geometry by T. • Estimate trace(THi) in local neighborhood of x. • Apply filtering for each trace(AijHj) in the vector trace(AH)

  29. Comparison Between Two Implementations

  30. Noise Removal and Image impainting

  31. Reconstruction, VisualizationSuperscale Images

More Related