We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

eess.SP

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Electrical Engineering and Systems Science > Signal Processing

Title: Towards Off-the-grid Algorithms for Total Variation Regularized Inverse Problems

Authors: Yohann de Castro (ICJ), Vincent Duval (CEREMADE, MOKAPLAN), Romain Petit (CEREMADE, MOKAPLAN)
Abstract: We introduce an algorithm to solve linear inverse problems regularized with the total (gradient) variation in a gridless manner. Contrary to most existing methods, that produce an approximate solution which is piecewise constant on a fixed mesh, our approach exploits the structure of the solutions and consists in iteratively constructing a linear combination of indicator functions of simple polygons.
Comments: For the short conference version see arXiv:2104.06706v2. For the long journal version see arXiv:2104.06706v3
Subjects: Signal Processing (eess.SP); Numerical Analysis (math.NA); Optimization and Control (math.OC)
Cite as: arXiv:2104.06706 [eess.SP]
  (or arXiv:2104.06706v3 [eess.SP] for this version)

Submission history

From: Romain Petit [view email]
[v1] Wed, 14 Apr 2021 08:56:09 GMT (1159kb,D)
[v2] Thu, 4 Nov 2021 13:00:14 GMT (1455kb,D)
[v3] Tue, 16 Nov 2021 17:20:34 GMT (1094kb,D)

Link back to: arXiv, form interface, contact.