Person-recognition using Time-Of-Flight Cameras
Department: Electrics/Electronics & Software
Diplomarbeit Ref.Nr. E_071
Advanced driver assistance and automated driving systems open up new middle-term perspectives in mobility and are therefore currently one of the major goals in the development of future vehicles. A new type of imaging sensor – the so called Time-of-Flight Camera – offers better opportunities to monitor the surroundings of a vehicle. Since ToF measures directly a depth image, the computing effort needed to generate a 3d map of the scene is reduced significantly.
For the EU funded project IoSense, VIF – in collaboration with the European partners – is currently equipping a demonstrator vehicle with ToF cameras. Initially common of-the-shelf computing hardware will be used for the evaluation of the depth images. Later on the functionality will be ported on dedicated embedded hardware components.
Goal of this particular master thesis is the recognition of human shapes within depth images. For 2D images several algorithms for person recognition are available, both commercially and open source. In context of this work these algorithms should be evaluated and modified accordingly to cope with depth images instead.
Your duties and responsibilites
- State-of-the-art survey (methods for person recognition)
- Generation of datasets used for training and testing the algorithms
- Evaluation of existing algorithms / libraries / toolboxes for person recognition (Matlab/C++)
- Porting/Implementation of person recognition algorithm on target hardware (C/C++)
- Testing of person detector within demonstrator vehicle
What we expect from you:
- Studies Telematics, Technical Informatics, Electrical Engineering, Mathematics or similar
- Basic knowledge C/C++ and Matlab beneficial
- Independent and conscientious way of working
What we offer:
- Proper supervision in the handling of this future related topic “automated driving” and collaboration with an international semiconductor supplier
- Collaboration with a committed, young and dynamic team
- Paid Master thesis
We look forward to receiving your details; please use our online contact form to send your application to VIRTUAL VEHICLE.Apply