CKSRI Seminar Series 2024 “Towards Holistic Scene Understanding for Autonomous Driving"
ABSTRACT:
This seminar has invited Prof. Kailun Yang, professor at the School of Robotics and the National Engineering Research Center of Robot Visual Perception and Control Technology, Hunan University to discuss towards holistic scene understanding for autonomous driving.
Holistic scene understanding of the driving surroundings is a prerequisite in safety-critical autonomous vehicle systems. To enable holistic scene understanding, multi-dimensional information should be sensed, fused, and exploited.
This talk delivers several works aiming to overcome the limits in sensing field of view, annotation deficiency, cross-temporal fusion, cross-modal fusion, and human-computer interaction towards holistic scene understanding. Addressing these challenges, recent solutions and benchmarks of panoramic scene segmentation, data-efficient occupancy estimation, vectorized HD map construction, arbitrary multimodal fusion, and referring scene understanding will be introduced.
SPEAKER: Prof. Kailun Yang is a professor at the School of Robotics and the National Engineering Research Center of Robot Visual Perception and Control Technology, Hunan University (HNU). He was a postdoctoral researcher at the Computer Vision for Human-Computer Interaction (CV:HCI) lab, Karlsruhe Institute of Technology (KIT). Prior to that, he received his Ph.D. degree from the State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang University (ZJU). His research interests include computer vision, computational optics, and their applications for autonomous driving and navigation assistance systems for vulnerable road users. He serves as an associate editor for IEEE Robotics and Automation Letters. His works reach top-10 popular article list in IEEE Transactions on Intelligent Transportation Systems, a top-cited paper in IEEE ICIP 2019, best paper award finalist in human-robot interaction in IEEE ICRA 2024, and a best paper award at 2021 IEEE Intelligent Symposium (IV).