User Tools

Site Tools


Dejan Pangercic


Technische Universität München

Department of Computer Science
Informatik 9
Karlstrasse 45 (3rd Floor)
80333 Munich

Tel: +49-89-289-26907
Fax: +49-89-289-26913
Office: CCRL II, 3002
Mail: dejan.pangercic_(at)_gmail_._com

We are looking for the students to work on the following topics as within their B.Sc., M.Sc, Guided Research projects:


Since January 2012 I have been working with Barcoo, a company from Berlin that is applying web mining for barcode applications (e.g. linking products in stores with the independent consumer information). I have a preliminary implementation of the barcode scanning and querying available here, please drop me an email for the details.


I have started my PhD at the Intelligent Autonomous Systems Group in 2009 under the supervisory of prof. Michael Beetz, PhD. My main research interest lies in semantic mapping, perception of everyday objects and scenes, mobile manipulation and system integration. I am one of the driving forces behind TUM's PR2 Beta Program proposal called CRAM and I spent summer of 2011 researching at Bosch RTC in Palo Alto, CA.

I am an active member of IEEE and RAS societies and regularly submit and publish to following conferences: IROS, ICRA, ICAR, Humanoids, RSS. I actively develop and contribute code to Robot Operating System (ROS), open-source, meta-operating system for personal robots. Please find my packages on server.

I obtained my B.Sc. and M.Sc. degrees at the Faculty for Electrical Engineering, UL and Department of Electrical Engineering and Information Technology, TUM respectively. Any additional information about me, including a complete CV, is available upon request.


Slides available from my talk in the Clutter 12 workshop at RSS 2012.
Read more news...


Semantic Mapping Scene Classification Object Recognition
Mobile Manipulation Cognitive Factory Robotic Demonstrations

* Courtesy: Moritz Tenorth
+ Courtesy: Zoltan-Csaba Marton


I use ROS in my research and thus integrate and release all of my code in form of ROS packages. Following list also includes packages I worked on together with my collaborators.

I am occasionally also contributing to Point Cloud Library.


In the past I have co-organized the following events:



There was a myriad of excellent undergrad colleagues that I got the chance to work together with. Below list is the list with their assignments and final reports.

  • Object categorization from models from WWW using Hough Voting by Vladyslav Usenko (together with Zoltan-Csaba Marton and Oscar Martinez Mozos) (still WIP)
  • RGB + D + High Level Feature based point cloud registration by Ross Kidson [ paper ] (still WIP)
  • Interactive Segmentation of Household Objects by Karol Hausman paper ] [ code ](Still WIP)
  • Detecting Unknown High-level Concepts in Semantic Mapping with Mobile Robots by Nikolaus Demmel (Co-supervised with Andrzej Pronobis) [ Final Presentation ]
  • Eye-in-Hand Modeling by Ronny Bismark [ Final Presentation ]
  • Barcode-based Object Recognition by Nacer Khalil [ Final Presentation] [ code ]
  • Android-based image processing by Rim Zahir [ Final Presentation]
  • Object detection and segmentation in cluttered scenes through perception and manipulation by Julius Adorf [ Final Presentation ]
  • Learning Where Objects Are - Organizational Principles in Human Environments by Martin Schuster (Co-supervised) [ Final Presentation ]
  • In-hand Object Modeling Center for Robotic Perception by Monica-Simona Opris [ Final Presentation ]
  • Implementation of Contracting Curve Density Algorithm for ROS by Shulei Zhu [ Final Presentation ] [ paper ]
  • Detection of Household Objects Using Projected Light Patterns by Florian Zacherl [ Final Presentation ] [ tutorial ]
  • Fast and Robust Object Detection in Household Environments Using DOT Descriptors and Vocabulary Trees with SIFT Descriptors by Vladimir Haltakov [ tutorial ] [ paper ]
  • Point Cloud Registration Framework by Hozefa Indorewala [ tutorial ] [ paper ]
  • Knowledge representation and learning algorithms utilising Robot Learning Language in the Cognitive Factory by Andreas Leha [ paper ]
  • Connecting High-Level Planning, Reasoning and Model-driven Vision into a Robotic System that Enables Everyday Manipulation Tasks by Rok Tavcar (Co-supervised) [ paper ]


people/pangercic.txt · Last modified: 2012/07/25 12:00 by pangerci