Martin Hirzer

I received an MSc degree (Dipl.-Ing.) and a PhD degree (Dr.techn.) in Telematics from Graz University of Technology in 2009 and 2014, respectively. Currently, I am a postdoctoral researcher at the Institute of Computer Graphics and Vision in the group of Professor Vincent Lepetit.

During my PhD studies, my research was focused on inter-camera person re-identification via (on-line) machine learning considering different feature representations and metric spaces. Additionally, I was involved in the Person ReID (ECV), the OUTLIER, the KiwiVision Cloud and the Mobi-Trick project.

I was also working on augmented reality (AR) technologies for CultAR, an EU FP7 project targeted at enhancing tourists’ experience when exploring cultural contents in urban environments. In particular, I was responsible for providing accurate estimates of the 6 degrees of freedom (DoF) camera pose of mobile devices such as smartphones or tablets to the overall system. In order to ensure real-time behavior despite the typically limited computational power of mobile platforms, this was realized by running a lightweight, relative rotation tracker (3 DoF) on top of an absolute, full 6 DoF localization service.

More recently, I worked on 3D camera tracking in urban environments based on 2.5D city maps and a semantic segmentation of the input frames. Specifically, by exploiting semantic image information, we were able to compensate for typical tracking inaccuracies such as drift.

Currently, my research is focused on indoor environments, in particular on the estimation of a room’s layout from a single RGB image. The goal is to find the boundaries between the floor, individual walls and the ceiling, which allows to infer the spatial layout. While this is very beneficial for many applications such as AR, indoor navigation, scene reconstruction and object detection, it is inherently challenging due to potentially severe occlusions, clutter, difficult lighting and large intra-class variance.

You can also find me on Google Scholar, ResearchGate, LinkedIn and XING. A full CV can be obtained upon request.

Publications

Efficient 3D Tracking in Urban Environments with Semantic Segmentation
Martin Hirzer, Clemens Arth, Peter M. Roth and Vincent Lepetit
In Proc. British Machine Vision Conf. (BMVC), 2017

Accurate Camera Registration in Urban Environments Using High-Level Feature Matching
Anil Armagan, Martin Hirzer, Peter M. Roth and Vincent Lepetit
In Proc. British Machine Vision Conf. (BMVC), 2017

Learning to Align Semantic Segmentation and 2.5D Maps for Geolocalization
Anil Armagan, Martin Hirzer, Peter M. Roth and Vincent Lepetit
In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2017

3D Localization in Urban Environments from Single Images
Anil Armagan, Martin Hirzer, Peter M. Roth and Vincent Lepetit
In Proc. OAGM/AAPR & ARW Joint Workshop (OAGM/AAPR & ARW), 2017

Semantic Segmentation for 3D Localization in Urban Environments
Anil Armagan, Martin Hirzer and Vincent Lepetit
In Proc. Joint Urban Remote Sensing Event (JURSE), 2017
Best Paper Award

Multiple Model Fitting by Evolutionary Dynamics
Michael Donoser, Martin Hirzer and Dieter Schmalstieg
In Proc. Int'l Conf. on Pattern Recognition (ICPR), 2014

Mahalanobis Distance Learning for Person Re-Identification
Peter M. Roth, Martin Hirzer, Martin Koestinger, Csaba Beleznai and Horst Bischof
In Person Re-Identification, pages 247–267, Springer, 2014

Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data
Csaba Beleznai, Michael Rauter, Martin Hirzer and Peter M. Roth
In Proc. Conf. of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF), 2013

Relaxed Pairwise Learned Metric for Person Re-Identification
Martin Hirzer, Peter M. Roth, Martin Koestinger and Horst Bischof
In Proc. European Conf. on Computer Vision (ECCV), 2012

Dense Appearance Modeling and Efficient Learning of Camera Transitions for Person Re-Identification
Martin Hirzer, Csaba Beleznai, Martin Koestinger, Peter M. Roth and Horst Bischof
In Proc. IEEE Int'l Conf. on Image Processing (ICIP), 2012

Person Re-Identification by Efficient Impostor-based Metric Learning
Martin Hirzer, Peter M. Roth and Horst Bischof
In Proc. IEEE Int'l Conf. on Advanced Video and Signal-Based Surveillance (AVSS), 2012

Large Scale Metric Learning from Equivalence Constraints
Martin Koestinger, Martin Hirzer, Paul Wohlhart, Peter M. Roth and Horst Bischof
In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2012

Multi-Cue Learning and Visualization of Unusual Events
Rene Schuster, Samuel Schulter, Georg Poier, Martin Hirzer, Josef A. Birchbauer, Peter M. Roth, Horst Bischof, Martin Winter and Peter Schallauer
In Proc. IEEE Int'l Workshop on Visual Surveillance (in conj. with ICCV), 2011

Person Re-Identification by Descriptive and Discriminative Classification
Martin Hirzer, Csaba Beleznai, Peter M. Roth and Horst Bischof
In Proc. Scandinavian Conf. on Image Analysis (SCIA), 2011

Saliency Driven Total Variation Segmentation
Michael Donoser, Martin Urschler, Martin Hirzer and Horst Bischof
In Proc. IEEE Int'l Conf. on Computer Vision (ICCV), 2009

An Automatic Hybrid Segmentation Approach for Aligned Face Portrait Images
Martin Hirzer, Martin Urschler, Horst Bischof and Josef A. Birchbauer
In Proc. Workshop of the Austrian Association for Pattern Recognition (OAGM/AAPR), 2009

Marker Detection for Augmented Reality Applications
Martin Hirzer
Technical Report, Graz University of Technology, Institute of Computer Graphics and Vision, ICG-TR-08/05, 2008

Theses

Combining Descriptive and Discriminative Information for Person Re-Identification
Martin Hirzer
PhD Thesis, Graz University of Technology, Faculty of Computer Science, 2014

Segmentation of Face Images
Martin Hirzer
MSc Thesis, Graz University of Technology, Faculty of Computer Science, 2008

Contact
image/svg+xml
Record not found: Person
Contents
image/svg+xml