Publica
Hier finden Sie wissenschaftliche Publikationen aus den FraunhoferInstituten. Energy minimization of discrete functions with higherorder potentials for depth map generation
:
Fulltext urn:nbn:de:0011n4265739 (3.3 MByte PDF) MD5 Fingerprint: ffa4417def41cfd65835c39022e74077 Created on: 3.1.2017 
 Institute of Electrical and Electronics Engineers IEEE: 23rd International Conference on Pattern Recognition, ICPR 2016 : Cancun, Mexico, December 48, 2016 Piscataway, NJ: IEEE, 2016 ISBN: 9781509048465 ISBN: 9781509048472 ISBN: 9781509048489 pp.23442349 
 International Conference on Pattern Recognition (ICPR) <23, 2016, Cancun> 

 English 
 Conference Paper, Electronic Publication 
 Fraunhofer IOSB () 
Abstract
Minimization of discrete energy functions considering higherorder potentials is a challenging yet an important problem. In this work, a threestep procedure will be presented and exemplified on a general problem related to the dense depth map computation from multiview configurations: Achieving a joint reconstruction of structure and semantics with piecewise planarity constraints. The three steps of the procedure are binarization, quadratization, and energy minimization. While the first and the third step are accomplished using procedures based on alphaexpansion and maxflow algorithms, respectively, we propose for the quadratization step a fast and simple module to reformulate the higherorder problem as a quadratic one. This module is based on edge statistics and is particularly useful for regular graphs and for third or fourthorder potentials.