eprintid: 8759
rev_number: 25
eprint_status: archive
userid: 1
dir: disk0/00/00/87/59
datestamp: 2008-11-06 14:04:01
lastmod: 2016-01-17 03:56:25
status_changed: 2012-08-14 15:26:45
type: preprint
metadata_visibility: show
creators_name: Lellmann, Jan
creators_name: Kappes, Jörg
creators_name: Yuan, Jing
creators_name: Becker, Florian
creators_name: Schnörr, Christoph
title: Convex Multi-Class Image Labeling by Simplex-Constrained Total Variation
ispublished: pub
subjects: 510
divisions: 708000
cterms_swd: Bildverarbeitung
cterms_swd: Konvexe Optimierung
cterms_swd: Diskrete Optimierung
cterms_swd: Funktion von beschränkter Variation
abstract: Multi-class labeling is one of the core problems in image analysis. We show how this combinatorial problem can be approximately solved using tools from convex optimization. We suggest a novel functional based on a multidimensional total variation formulation, allowing for a broad range of data terms. Optimization is carried out in the operator splitting framework using Douglas-Rachford Splitting. In this connection, we compare two methods to solve the Rudin-Osher-Fatemi type subproblems and demonstrate the performance of our approach on single- and multichannel images.
abstract_translated_lang: eng
date: 2008
date_type: published
id_scheme: DOI
id_number: 10.11588/heidok.00008759
portal_cluster_id: p-iwrpp
portal_order: 08759
ppn_swb: 1647466601
own_urn: urn:nbn:de:bsz:16-opus-87593
language: ger
bibsort: LELLMANNJACONVEXMULT2008
full_text_status: public
series: IWR-Preprints
citation:   Lellmann, Jan ; Kappes, Jörg ; Yuan, Jing ; Becker, Florian ; Schnörr, Christoph  (2008) Convex Multi-Class Image Labeling by Simplex-Constrained Total Variation.  [Preprint]     
document_url: https://archiv.ub.uni-heidelberg.de/volltextserver/8759/1/lellmann08multiclasstv1.pdf