||Dr. Xin Yuan (Bell Labs)
||Prof. Jian Wang (Fudan University)
||14:00 – 16:00, July 15, 2019
||Zibin N201, Fudan University
||Snapshot compressive imaging (SCI) refers to compressive imaging systems where multiple frames are mapped into a single measurement, which has been used in video, hyperspectral imaging, depth imaging, polarization imaging, microscopy and x-ray imaging. Though the hardware has been investigated for more than a decade, the theoretical guarantees have recently been developed. This talk will first discuss the performance bounds of SCI systems in the information theory framework. The state-of-the-art optimization based inverse algorithm (DeSCI) will be presented; inspired by deep learning, we will present λ-net, which is developed for the inverse problem of hyperspectral compressive imaging. Following this, diverse applications of SCI will be discussed.
||Dr. Xin Yuan is currently a video analysis and coding lead researcher at Bell Labs, Murray Hill, NJ, USA. Prior to this, he had been a Post-Doctoral Associate with the Department of Electrical and Computer Engineering, Duke University from 2012 to 2015, where he was working on compressive sensing and machine learning. He develops compressive sensing techniques for high-dimensional imaging with applications to videos, hyperspectral, microscopy and x-ray imaging. Before joining Duke, Dr. Yuan obtained his B.Eng and M.Eng from Xidian University in 2007 and 2009, respectively, and his Ph.D. from the Hong Kong Polytechnic University in 2012. Two papers he coauthored won the best paper award in the Computational Optical Sensing and Imaging (COSI) conferences in 2013 and 2014, on video compressive sensing and on depth compressive sensing cameras, respectively.