Multi-Resolutive Sparse Approximations of d-Dimensional Data (Articolo in rivista)

  • Multi-Resolutive Sparse Approximations of d-Dimensional Data (Articolo in rivista) (literal)
  • 2013-01-01T00:00:00+01:00 (literal)
  • 10.1016/j.cviu.2012.10.012 (literal)
Alternative label
  • Giuseppe Patanè (2013)
    Multi-Resolutive Sparse Approximations of d-Dimensional Data
    in Computer vision and image understanding (Print); Elsevier Inc., San Diego (Stati Uniti d'America)
  • Giuseppe Patanè (literal)
Pagina inizio
  • 418 (literal)
Pagina fine
  • 428 (literal)
  • 117 (literal)
  • 4 (literal)
  • ISI Web of Science (WOS) (literal)
  • Scopu (literal)
  • Consiglio Nazionale delle Ricerche, Istituto di Matematica Applicata e Tecnologie Informatiche, Genova, Italy (literal)
  • Multi-Resolutive Sparse Approximations of d-Dimensional Data (literal)
  • This paper proposes an iterative computation of sparse representations of functions defined on Rd, which exploits a formulation of the sparsification problem equivalent to Support Vector Machine and based on Tikhonov regularization. Through this equivalent formulation, the sparsification reduces to an approximation problem with a Tikhonov regularizer, which selects the null coefficients of the resulting approximation. The proposed multi-resolutive sparsification achieves a different resolution in the approximation of the input data through a hierarchy of nested approximation spaces. The idea behind our approach is to combine a smooth and strictly convex approximation of the l1-norm with Tikhonov regularization and iterative solvers of linear/non-linear equations. Firstly, the iterative sparsification scheme is introduced in a Reproducing Kernel Hilbert Space with respect to its native norm. Then, the sparsification is generalized to arbitrary function spaces using the least-squares norm and radial basis functions. Finally, the discrete sparsification is derived using the eigendecomposition and the spectral properties of sparse matrices; in this case, the computational cost is O(nlogn), with n number of input points. Assuming that the data is supported on a (d ? 1)-dimensional manifold, we derive a variant of the sparsification scheme that guarantees the smoothness of the solution in the ambient and intrinsic space by using spectral graph theory and manifold learning techniques. Finally, we discuss the multiresolutive approximation of d-dimensional data such as signals, images, and 3D shapes. (literal)
Prodotto di
Autore CNR
Insieme di parole chiave

Incoming links:

Autore CNR di
Editore di
Insieme di parole chiave di