Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB
Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB
Kameraposition mit Kamerabild relativ zum 3D-Modell der Umgebung
Gegenstand der Arbeiten ist die Erkennung und Übergabe von Objekten in den bildhaften Daten vernetzter, mobiler Sensoren. Die untersuchten Technologien reichen dabei von der ansichtenunabhängigen Objektbeschreibung über die Registrierung von Sensorbildern in dreidimensionale Kontextdaten bis hin zur bandbreitensparenden Übertragung der Objektinformationen.

Object Recognition in Sensor Networks

Group Description:

The main topic of this field is the automatic detection and hand-over of objects within the data streams of a network of sensors. This includes the view-invariant description of objects, the multi-modal registration of video and laser sensor data, and the low bandwidth transfer of object information within the sensor network.

 

Team:

Dr. Christoph Bodensteiner
Timo Breuer
Dr. Sebastian Bullinger
Florian Fervers
Valentin Schorb
Gregor Stachowiak
Valentin Wagner

Projects 

  • Image based Object Localization
    The project aims for generic and high precision object geo-localization methods using previously acquired, large scale 3-D imagery.
  • Reconstruction of Moving Objects in Monocular Video Data.
    This project focuses on the reconstruction of the three-dimensional shape of moving objects in monocular video data using semantic information.
  • 3D Vehicle Trajectory Reconstruction in Monocular Video Data
    The determination of consistent object trajectories requires the definition of additional motion constraints. This project analyzes and evaluates suitable constraints.
  • The MODISSA testbed
    MODISSA (Mobile Distributed Situation Awareness) is the IOSB's realization of an experimental platform for hardware evaluation and software development in the contexts of automotive safety, security, and military applications.

List of Publications:

  • Bodensteiner, Bullinger, S., C., Arens, M.: Multispectral Matching using Conditional Generative Appearance Modeling. AVSS, Auckland, New Zealand, 2018.
  • Bullinger, S., Bodensteiner, C., Arens, M.: Monocular 3D Vehicle Trajectory Reconstruction Using Terrain Shape Constraints. ITSC, Maui, USA, 2018.
  • Bullinger, S., Bodensteiner, C., Arens, M., Stiefelhagen, R.: 3D Vehicle Trajectory Reconstruction in Monocular Video Data Using Environment Structure Constraints. ECCV, Munich, Germany, 2018.
  • Bullinger, S., Bodensteiner, C., Arens, M.: Instance Flow Based Online Multiple Object Tracking. ICIP, Bejing, China, 2017.
  • Bullinger, S., Bodensteiner, C., Arens, M.: Moving Object Reconstruction in Monocular Video Data Using Boundary Generation. ICPR, Cancún, Mexico, 2016.
  • Bodensteiner, C., Bullinger, S., Arens, M.: Single Frame Based Video Geo-Localisation using Structure Projection. ICCVW, Santiago de Chile, Chile, 2015.
  • Lemaire S., Bodensteiner C., Arens M.: „Mobile device geo-localization and object visualization in sensor networks", In: Proc. of the SPIE, Vol. 9250, 2014.
  • Breuer T., Bodensteiner C., Arens M.: „Low-cost commodity depth sensor comparison and accuracy analysis", In: Proc. of the SPIE, Vol. 9250, 2014.
  • Hesse N., Bodensteiner C., Arens M.: „Performance evaluation of image-based location recognition approaches based on large-scale UAV imagery", In: Proc. of the SPIE, Vol. 9250, 2014.
  • Lemaire S., Bodensteiner C., Arens M.: „High precision object geo-localization and visualization in sensor networks", In: Proc. of the SPIE, Vol. 8899, 2013.
  • Bodensteiner C.: A Situation Awareness and Information Handover System for Small Unit Operations, Trilateral-Workshop (SE-NL-DE) Team Situation Awareness in Small Unit Operations, Stockholm, 2011.
  • Bodensteiner C., Hübner W., Jüngling K., Solbrig P., Arens M.: Monocular Camera Trajectory Optimization using LiDAR Data, Computer Vision in Vehicle Technology (CVVT), ICCV, Barcelona, 2011.
  • Bodensteiner, C., Hebel, M., Arens, M.: Accurate Single Image Multi-Modal Camera Pose Estimation. ECCV – RMLE, Crete, 2010.
  • Bodensteiner, C., Huebner, W., Juengling, K., Mueller, J., Arens, M.: Local Multi-Modal Image Matching Based on Self-Similarity, ICIP, Hongkong, 2010.