Discrete Mumford-Shah on graph for mixing matrix estimation

Abstract : The discrete Mumford-Shah formalism has been introduced for the image denoising problem, allowing to capture both smooth behavior inside an object and sharp transitions on the boundary. In the present work, we propose first to extend this formalism to graphs and to the problem of mixing matrix estimation. New algorithmic schemes with convergence guarantees relying on proximal alternating minimization strategies are derived and their efficiency (good estimation and robustness to initialization) are evaluated on simulated data, in the context of vote transfer matrix estimation.
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01948570
Contributor : Marion Foare <>
Submitted on : Monday, May 27, 2019 - 10:17:27 AM
Last modification on : Wednesday, May 29, 2019 - 1:06:05 AM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01948570, version 2

Citation

Yacouba Kaloga, Marion Foare, Nelly Pustelnik, Pablo Jensen. Discrete Mumford-Shah on graph for mixing matrix estimation. 2019. ⟨hal-01948570v2⟩

Share

Metrics

Record views

36

Files downloads

63