Paul Swoboda

E-Mail: swoboda (at) math.uni (minus) heidelberg.de

Scientific Interests

Publications

Maximum Persistency via Iterative Relaxed Inference with Graphical Models
Alexander Shekhovtsov, Paul Swoboda, Bogdan Savchynskyy, CVPR 2015. pdf bib
project page: Persistency via Iterative Relaxed Inference.

Partial Optimality by Pruning for MAP-inference with General Graphical Models
Paul Swoboda, Bogdan Savchynskyy, Jörg H. Kappes and Christoph Schnörr, CVPR 2014. pdf bib
Accompanying code: Persistency by Pruning.

Global MAP-Optimality by Shrinking the Combinatorial Search Area with Convex Relaxation
Bogdan Savchynskyy, Jörg H. Kappes, Paul Swoboda and Christoph Schnörr, NIPS 2013. pdf bib

Variational Image Segmentation and Cosegmentation with the Wasserstein Distance
Paul Swoboda and Christoph Schnörr, EMMCVPR 2013. pdf bib

Partial Optimality via Iterative Pruning for the Potts Model.
Paul Swoboda, Bogdan Savchynskyy, Jörg H. Kappes and Christoph Schnörr, SSVM 2013. pdf. bib. Accompanying talk: pdf

Convex Variational Image Restoration with Histogram Priors.
Paul Swoboda and Christoph Schnörr, SIAM J.~Imag.~Sci., pdf bib

Code

Region Covariance Descriptors implemented in pure Matlab as described in "Region Covariance: A Fast Descriptor for Detection and Classification" by Oncel Tuzel, Fatih Porikli, and Peter Meer, ECCV 2006: Region covariance descriptor.

Contact

Paul Swoboda
ipa - Image and Pattern Analysis Group
Department of Mathematics and Computer Science
University of Heidelberg
Speyerer Str. 6, room G 2.07
69115 Heidelberg, Germany

Voice: (+49) 6221 - 54 5272