Pose Invariant Deformable Shape Priors Using L1 Higher Order Sparse Graphs

Abstract : In this paper we propose a novel method for knowledge-based segmentation. We adopt a point distribution graphical model formulation which encodes pose invariant shape priors through L1 sparse higher order cliques. Local shape deformation properties of the model can be captured and learned in an optimal manner from a training set using dual decomposition. These higher order shape terms are combined with conventional visual ones aiming at maximizing the posterior segmentation likelihood. The considered graphical model is optimized using dual decomposition and is used towards 2D (computer vision) and 3D object segmentation (medical imaging) with promising results.
Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download

https://hal.inria.fr/hal-00856978
Contributor : Bo Xiang <>
Submitted on : Monday, September 2, 2013 - 5:17:42 PM
Last modification on : Thursday, February 7, 2019 - 5:29:17 PM
Long-term archiving on : Tuesday, December 3, 2013 - 6:10:09 AM

File

isvc382.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00856978, version 1

Citation

Bo Xiang, Nikos Komodakis, Nikos Paragios. Pose Invariant Deformable Shape Priors Using L1 Higher Order Sparse Graphs. 9th International Symposium on Visual Computing - ISVC 2013, Jul 2013, Rethymnon, Greece. ⟨hal-00856978⟩

Share

Metrics

Record views

599

Files downloads

389