UC Berkeley DeepDrive and Its Driving Data Efforts

The self-driving boom continues. In June 2018, the University of California, Berkeley's Institute of Industrial Intelligence (BAIR), one of the most authoritative research institutions in the field of autonomous driving, released the largest driving video dataset BDD100K, which received high attention from the driving industry and the academic community. DTalk episode 4 has the honor to invite one of the BDD100K project members, Berkeley DeepDrive (BDD) researcher Fisher Yu to bring you the latest and most cutting-edge self-driving technologies and computer vision research efforts of BDD. Fisher Yu will introduce the major achievements of BDD in the automatic driving perception algorithm and the driving video dataset and explain some related technologies. The sharing session will be full of incredible stuff that should not be missed!

Speaker: Dr. Fisher Yu
Fisher Yu is a postdoctoral researcher at UC Berkeley, working with Trevor Darrell. He pursued his Ph.D. degree at Princeton University, advised by Thomas Funkhouser. During his Ph.D. study, he also collaborated extensively with Vladlen Koltun at Intel and Jianxiong Xiao at Princeton. His research interest lies in representation learning for image recognition, Internet-scale visual understanding, and vision-based autonomous driving system. He is the co-organizer of multiple CVPR workshops on autonomous driving and challenges for several years and he has served as reviewers for major computer vision, machine learning, and robotics conferences such as CVPR, ECCV, NIPS, ICML, ICRA, and IROS. He leads the construction and annotation of BDD100K (bdd-data.berkelkey.edu) dataset, and the development of Scalable (www.scalabel.ai), an open-sourced versatile visual data annotation system.

Guest Speaker: Dr. Mianwei Zhou, Research Scientist @ PlusAI
Mianwei Zhou, a founding team member and research scientist at PlusAI. He got his computer science bachelor degree from Shanghai Jiaotong University ACM Honor Class and a Ph.D. degree from University of Illinois Urbana Champaign. Dr. Mianwei Zhou focuses on data mining/machine learning areas, and have publications on top international conference such as KDD, SIGMOD. In PlusAI, He leads the modeling team and worked on various research problems including computer vision, slam/localization, and model predictive control.

Agenda:18:30-20: 30 pm 29th August 2018

18:30-19:00 Check-In

19:00-19:30 Speach by Fisher Yu

19:30-20:15 Fireside Chat: Fisher Yu and Mianwei Zhou

20:15-20:30 Q&A