3D Perception and Prediction for Large-scale Traffic Scenarios

Publisher:闻天明Release Time:2020-12-08Number of visits:11

Speaker:   Prof. Yuexin Ma

Time:       16:30-17:30, Dec. 11

Location:  SIST 1C 101

Host:        SIST

 

Abstract

Autonomous driving is a significant task that has the potential to impact peoples’ day-to-day lives. It is also a difficult task that requires high-quality devices and mature algorithms to guarantee the safety and efficiency. Challenges arise in complex urban environments, where the traffic consists of heterogeneous traffic-agents, including cars, trucks, buses, bicycles, pedestrians, etc. 3D perception and prediction, including 3D segmentation, 3D detection, behavior and trajectory prediction, are important tasks for autonomous driving system because they can tell autonomous vehicles what and where neighboring objects are and where they will go. We proposed efficient and reliable algorithms to solve these problems and achieved state-of-the-art performance.

 

Bio

Yuexin Ma is currently an assistant professor in SIST, ShanghaiTech University. Before that, she was a research assistant professor in Hong Kong Baptist University. She got PhD degree from the department of Computer Science of the University of Hong Kong. Her research interests include computer vision, autonomous driving, and computer graphics. She has published papers in Science Robotics, ECCV, AAAI, IJCAI, AAMAS, TOG, SIGGRAPH ASIA, etc. She got the Hong Kong PhD Fellowship Award (HKPFS) in 2015. She visited French Institute for Research in Computer Science and Automation (INRIA) in 2016 and the University of North Carolina at Chapel Hill (UNC) in 2017.

 

SIST Seminar 202009