CO-LOCATED EVENTS
NextPrevious

Session

3 | Sensor Fusion Café

Tuesday, September 30

11:00 AM - 03:00 PM

Live in Berlin

Less Details

  • How can simulated data address the lack of rare accident-prone traffic and weather events in training models?
  • Should sensor fusion systems be able to selectively deactivate sensors to save power, and under what conditions would they reactivate?
  • How should sensor fusion adapt to different risk levels, reaction times, and safety distances in varied driving environments?
  • Should inference systems adapt to changes in various driving conditions, or rely on the extensive training with diverse datasets that cover these conditions?
  • How can sensor fusion better handle conflicting or missing data from sources like X2V or V2V communication?
PE
Workshop

Speaker

Wilfried Philips

Professor Image Processing and Interpretation, Ghent University

Wilfried Philips is senior full professor at Ghent University, where he heads the research group IPI (Image Processing and Interpretation). He also leads the sensor fusion research within the research organization imec and is co-founder of Ghent University's UAV research center and UGent's research valorization consortium iKNOW which fosters and develop research partnerships and collaborations with industry, and stimulates spin off creation and IP transfer. Wilfried Philips's main research interests are in the domains of image and video quality improvement, real-time computer vision and multi-modal sensor fusion of regular, hyper-spectral and thermal camera, radar and LiDaR. Road user detection and tracking for Drive Assistance and autonomous driving, safety in factories and at sea and advanced multimodel mapping from UAVs and remote sensing are the most important application areas. IPI has a rich history of both academic research and cooperation with industry, featuring over 100 projects with industry, and many EU projects. The research group has created several spin-offs, including Pozyx (industrial real time location tracking) and Senso2Me (Internet of Things solutions for elderly care).

The Pop in Your Job – What drives you? Why do you love your job?
At IPI we like creating solutions to real-world problems, specifically difficult sensing and scene analysis problems. As an academic research group we like to develop fundamental concepts, such as specific fusion architectures or restoration algorithms. However, we also like those concepts to find practical use. We have therefore cooperated a lot with industry, thus learning about a wide variety of sensing sensor fusion and computer vision problems and appreciating industrial requirements such as the need to achieve specific accuracy levels, processing speeds, and cost levels. This has not only lead to applied research on very concrete topics with local companies, but also to more elaborate research programs with multinationals. This cooperation has also helped us evaluate the value and restrictions of our research in real-world applications, including on real-life data captured in real-life traffic.

Company

Ghent University

Ghent University is a top 100 university founded in 1817. It is the highest ranking Belgian University in the well-known Shanghai academic ranking. Our 11 faculties offer a comprehensive range of over 200 master programs to over 50 000 students in Belgium, but also at the Incheon Global campus in Korea. Ghent University is quite active in European research programs and has attracted over 140 ERC grants. We are also quite active in technology transfer, resulting in 84 spin-offs and over 1 000 patents. We are ranked high up in the top 5 of Digital Scale-Ups according to Omar Mohout's ranking. Ghent University's Faculty of Engineering and Architecture cooperates closely with the world-leading research center imec, the world's leading independent nanoelectronics R&D hub. The combination of infrastructure, and partner network, including close cooperation with universities enables breakthroughs towards microchips that are smaller, faster, more affordable, and more sustainable. We combine that nanotech expertise, with data analysis, sensor fusion and AI to create and support applications for a smarter, better future. Automotive applications are one of the important application areas.

NextPrevious