SS17 - Probabilistic RGBD Data Fusion

In contrast to a single sensor,  the  combination of multiple cameras brings several advantages, including   simultaneous coverage of a large environment, increased resolution,   redundancy, and robustness against occlusion. However, together with  these great benefits a variety of challenges  arise: synchronization,  calibration, registration, multi-sensor fusion,  large amounts of data,  interference issues, and last but not least,  sensor-specific stochastic  and set-valued uncertainties. 

This  Special Session addresses fundamental techniques, recent  developments  and future research directions in the field of  probabilistic RGBD data  fusion. 

Topics of interest 

•  Methodologies for probabilistic RGBD data fusion: Bayesian inference, nonlinear filtering, random sets, data association 
•  Sensor models 
•  Sensor management, calibration, registration, synchronization 
•  Sensors: Radar devices, laser rangefinders, RGB cameras, depth cameras 
•  Applications: Surveillance, telepresence, motion capturing, 3D reconstruction, robotics, medicine, biology, computer vision 
•  Case studies: Benchmark scenarios, performance measures 

Keywords 

RGBD sensor (networks), point cloud fusion, object tracking, calibration 

Special Session Organizers

• Uwe D. Hanebeck, Karlsruhe Institute of Technology (Germany)
• Florian Faion, Karlsruhe Institute of Technology (Germany)
• Antonio Zea, Karlsruhe Institute of Technology (Germany)

Special Session Contact

• Uwe D. Hanebeck ()
• Florian Faion ()
• Antonio Zea ()