Speaker: Yotam Gil

M.Sc. student under the supervision of Prof. Raja Giryes

Wednesday, February 19^th, 2020 at 15:00

Room 011, Kitot Bldg., Faculty of Engineering

MonSter - Awakening the Mono in Stereo

Abstract

A novel stereo imaging system is presented, achieving enhanced performance over traditional stereo cameras, as well as self calibration abilities.

Stereo imaging is the most common passive method for producing reliable depth maps, however it suffers from two cardinal issues: limited depth range due to disparity resolution and sensitivity to extrinsic calibration. In this thesis, we offer a framework to overcome these limitations with the help of previously ignored information from each of the stereo images. First, we show how a stereo depth-map can be improved by equipping one of the stereo cameras with a phase-coded mask tuned for a range of depths in which the stereo struggles; in these ranges we use the monocular depth map as another source of depth information. This leads to a depth map fusion method that improves the original stereo depth map accuracy by 10%. Second, we present a novel online self-calibration approach, which makes use of both the stereo and monocular depth maps to find the transformation required for calibration by enforcing consistency between both maps. an extrinsic calibration is a crucial step for every stereo based system; despite all the advancements in the field, most calibrations are still done by the same tedious method of a checkerboard target. Monocular based depth estimation methods do not require extrinsic calibration but generally achieve inferior depth accuracy. The proposed method works in a closed-loop and exploits the pre-trained networks' global context, and thus avoids feature matching and outliers issues. In addition to presenting our method using an image-based monocular depth estimation method, which can be implemented in most systems without additional changes, we also show that having a phase-coded aperture mask leads to even better and faster convergence. We demonstrate our method on road scenes from the KITTI vision benchmark and real-world scenes using our prototype camera.

(The talk will be given in English)

Speaker:     Prof. Yoram Bresler

Coordinated Science Laboratory and the Department of ECE, 

University of Illinois at Urbana-Champaign
 

Monday, February 10^th, 2020
15:00 - 16:00

Room 011, Kitot Bldg., Faculty of Engineering

Faster Guaranteed GAN-based recovery in Linear Inverse Problems

Abstract

A Generative Adversarial Network (GAN) trained to model the prior of images has been shown to perform better than sparsity-based regularizers in ill-posed inverse problems. We describe an approach along these lines, with some modifications and refinements, with the following features: (1) on a given class of images, it addresses different linear inverse problems without re-training the neural network; (2) it accelerates the computation substantially as compared to previous GAN-based methods; and (3) it comes with a recovery guarantee. Experiments on several inverse problems demonstrate substantial speedup over earlier GAN-based recovery methods, along with better accuracy.
 

Short Bio

Yoram Bresler received the B.Sc. and M.Sc.  from the Technion–Israel Institute of Technology, and the Ph.D. from Stanford University all in electrical engineering. He is currently a Founder Professor of Engineering with the Departments of Electrical and Computer Engineering and Bioengineering, and the Coordinated Science Laboratory at the University of Illinois at Urbana–Champaign. He is also the founding President and the Chief Technology Officer at InstaRecon, Inc., commercializing breakthrough technology for tomographic reconstruction developed in his academic research. His research interests include machine learning and statistical signal processing and their applications to inverse problems in imaging, including compressed sensing, computed tomography, and magnetic resonance imaging.
Dr. Bresler was a Faculty Fellow at the National Center for Super Computing Applications in 2006. He is a recipient of the 1991 NSF Presidential Young Investigator Award, the Technion Fellowship in 1995, and the Xerox Senior Award for Faculty Research in 1998. He was named as the University of Illinois Scholar in 1999, and was appointed as an Associate at the Center for Advanced Study of the University in 2001. He has served on the editorial board of several journals, including the IEEE TSP, the IEEE JSTSP, Machine Vision and Applications, the SIAM Journal on Imaging Science, and on various committees of the IEEE.
Dr. Bresler is a fellow of the IEEE and the AIMBE. His papers have received 4 best journal paper awards, two of these with his students. In 2016, he was appointed IEEE SPS Distinguished Lecturer.‏