Abstract

In this talk I will present some of our recent work on lensless imaging using programmable optical  elements. Existing methods for lensless imaging can recover the depth and intensity of the scene, but they  require solving computationally-expensive inverse problems. I will start with a discussion on how we can  recover RGBD images using lensless cameras. I will then discuss how we can simplify the 3D recovery  problem using a small number of measurements with a shifting or programmable mask. In a nutshell, the  recovery process involves diagonalization of convolution operators in the Fourier domain and solving  multiple small least-squares problems. I will also present a design template for optimizing the mask  patterns with the goal of improving depth estimation and experimental results from a camera prototype.  Finally, I will present some of our recent results on using coded illumination with lensless imaging that  significantly improves the quality of reconstructed images.

 

Bio

M. Salman Asif is an Assistant Professor in the Department of Electrical and Computer Engineering at the  University of California, Riverside. Dr. Asif received his B.Sc. degree from the University of Engineering  and Technology, Lahore, Pakistan, and his M.S.E.E and Ph.D. degrees from the Georgia Institute of  Technology, Atlanta, Georgia. He worked as a Senior Research Engineer at Samsung Research America,  Dallas and as a Postdoctoral Researcher at Rice University. He is a recipient of NSF CAREER Award,  Google Faculty Award, UC Regents Faculty Fellowship Award, and Hershel M. Rich Outstanding  Invention Award. His research interests include computational imaging, compressive sensing, computer  vision, and machine learning.

 

Host

Professor Stanley Chan