Home  |  Centre for Mathematical Sciences  |  LTH  |  LU
Title: Curvature Regularity for Multi-Label Problems - Standard and Customized Linear Programming
Full text: PDF
Authors: Schoenemann, Thomas and Kuang, Yubin and Kahl, Fredrik
Year: 2011
Document Type:Conference Paper
Conference: EMMCVPR
Conference location: St Petersburg
Status: Published
Refereed: Yes
BibTeX item:BibTeX
Abstract: We follow recent work by Schoenemann et al. [25] for expressing curvature regularity as a linear program. While the original formulation focused on binary segmentation, we address several multi-label problems, including segmentation, denoising and inpainting, all cast as a single linear program. Our multi-label segmentation introduces a curvature Potts model" and combines a well-known Potts model relaxation [14] with the above work. For inpainting, we improve on [25] by grouping intensities into bins. Finally, we address the problem of denoising with absolute di erences in the data term. Furthermore, we explore alternative solving strategies, including higher order Markov Random Fields, min-sum di usion and a combination of augmented Lagrangians and an accelerated rst order scheme to solve the linear programs.




Questions: webmaster
Last updated: 2013-06-04

Centre for Mathematical Sciences, Box 118, SE-22100, Lund. Phone: 046-222 00 00