Abstract:

Autonomous vehicles rely on various sensors to perceive their surroundings and plan driving decisions accordingly. However, these sensors can only provide partial and biased measurements of the environment --e.g., due to occlusions or range limitations. To cope with these inherent limitations, sharing data between surrounding road agents is an emerging solution to build a more robust and ubiquitous representation of the environment that can be used for navigation, traffic monitoring, assisted driving, and much more.

In this talk, we will discuss how connected vehicles can benefit from computer vision to localize a swarm of cars and perform real-time augmented reality. This seminar will also address other topics related to recent developments in vehicular perception and the concept of smart-city, which both are shaping the future of autonomous driving.

Bio:

Francois Rameau is an Associate Research Professor at KAIST (South Korea). He received his Master's degree in Vision and Robotics from the University of Burgundy (France) in 2011. In the same institution, he completed his Ph.D. degree (about 3D vision for robotics) in December 2014 under Prof. Cedric Demonceaux and Prof. David Fofi.

Dr. Rameau joined KAIST (South Korea) in 2015, first as a post-doctoral researcher and then as Research Professor in 2017 under the KRF fellowship program.

His research interests include machine learning, 3D reconstruction, and perception for robotics. He is one of the pioneers in computer vision for connected vehicles, where he developed novel approaches, such as collaborative localization and augmented reality for assisted driving. Please refer to his personal webpage for more details: https://rameau-fr.github.io/ 

LOCATION:
 Zoom  - email events [at] cs.stonybrook.edu for access.