Internship Computer Vision with Event Camera

Internship Computer Vision with Event Camera at NLR

Job description

Are you interested in an interesting internship at NLR with a focus on computer vision and sensor processing?

What will you do?

We have identified several promising algorithms and applications of this technology which we would like to investigate further. These mainly concern algorithms for navigation purposes, but there is room for your own suggestions. We are looking for answers to questions such as:

• What is the current state of the art as for event-based SLAM?
• How well do available algorithms perform in comparison to frame-based SLAM?
• What are the challenges in terms of hardware setup?
• What are the opportunities in terms of fusion with different sensor modalities?
• What are the challenges in terms of calibration and initialisation?
• How can we integrate available algorithms into our own navigation methods?

What you bring?

We’re looking for a master WO-student with a background in Robotics, Computer Vision, Electrical Engineering, Applied Mathematics/Physics or similar, who:

• is enthusiastic about computer vision and sensor processing,
• has experience with ROS or ROS2 and programming, preferably in C++,
• has experience with image processing libraries such as OpenCV,
• is assertive, curious, motivated and able to work independently

What we offer?

A challenging and interesting internship project where you get the chance to lay your hands on an innovative, high-tech sensor: the event camera. The project is meant to have a duration of 3 months, but can be extended if wanted/necessary. Start date is as soon as possible, in consultation with the student. We offer an attractive internship allowance.

Your new working environment

Within the navigation community at the Netherlands Aerospace Centre (NLR), we research various methods for autonomous navigation of mobile platforms. A promising approach that has been globally widely studied is navigation based on a visual SLAM algorithm, in which SLAM stands for Simultaneous Localisation and Mapping. Experimental results at our facilities confirmed the promising characteristics of this type of navigation, but also highlighted some inherent weaknesses of the use of standard cameras. An ongoing research direction in the worldwide computer vision community is the utilisation of a recent innovation: event-based vision for SLAM. Event-based cameras are fundamentally different from regular (frame-based) cameras, leading to several benefits as a high dynamic range and no sensitivity to motion blur. The deployment of these emerging sensors for our use-cases requires further research.

What does this mean working at NLR?

Royal NLR has been the ambitious research organisation with the will to keep innovating for over 100 years. With that drive, we make the world of transportation safer, more sustainable, more efficient and more effective. We are on the threshold of breakthrough innovations. Plans and ideas start to move when these are fed with the right energy. Over 950 driven professionals work on research and innovation. From aircraft engineers to psychologists and from mathematicians to application experts.

For this position, an integrity test and the acquisition of a ‘certificate of no objection’ (VGB) are prerequisites for entry. More information on the VGB is available on the website of the Ministry of Justice.