Mining the Dynamics
- Thursday, January 12, 2017, 11:00 am - 12:00 pm PDTiCal
- 11th floor large conference room
This event is open to the public.
- AI Seminar
- Dr. Fei Xiong, of Amazon Inc.,
Abstract: Detecting and describing objects and their interactions in video sequences remain a very challenging task. This presentation introduces a set of tools combining elements from computer vision, dynamical system theory, optimization and machine learning for dynamic data association, classification and dimension reduction. The main idea driving the introduced methods is the fundamental fact that the underlying dynamics are the compact representations of the information encapsulated in the video data. The potential of these tools is also shown in the presentation through several challenging applications including tracking, manifold embedding of video data to recover missing data and forecast future measurements, and classification.
Biography: Fei Xiong received the B.S. and M.S. degree from Huazhong University of Science and Technology, in 2004 and 2007, respectively, and Ph.D. degree in Electrical Engineering from Northeastern University, Boston, in 2014. After graduation, he joined Amazon Inc. as the initial member of Amazon Go project. His current research interests include visual tracking and object recognition, video segmentation, manifold learning, and 3D reconstruction.