Speaker: Keren Rotker

M.Sc. student under the supervision of Prof. Alex Bronstein and Prof. Dafna Ben-Bashat

Wednesday, June 13^th, 2018 at 15:00

Room 011, Kitot Bldg., Faculty of Engineering

Over-Parameterized Models for Vector Fields with Application to Phase-Contrast MRI Data

Abstract

Vector fields arise in a variety of quantity measure and visualization techniques such as fluid flow imaging, motion estimation, deformation measures and color imaging, leading to a better understanding of physical phenomena. Recent progress in vector field imaging technologies has emphasized the need for efficient noise removal and reconstruction algorithms. A key ingredient in the success of extracting signals from noisy measurements is prior information. This prior knowledge can often be represented as a parameterized model. In this work, we extend the over-parameterization variational framework in order to perform model-based noise removal of vector fields. The over-parameterization methodology combines local modeling of the data with global model parameter regularization. By considering the vector field as a linear combination of basis vector fields and appropriate scale and rotation coefficients, the denoising problem reduces to a simpler form of coefficient recovery. We introduce two versions of the over-parameterization framework: total variation-based method and sparsity-based method, relying on the cosparse analysis model. We first demonstrate the efficiency of the proposed frameworks for two-

and three-dimensional vector fields with linear over-parameterization models. We then consider color images as vector fields and illustrate denoising via the new techniques. Finally, we address the problem of denoising magnetic resonance imaging (MRI) vector field data. Advances in medical imaging technologies have led to new modalities such as flow-sensitive magnetic resonance imaging (phase-contrast MRI) which allows the acquisition of blood flow velocities with a volumetric coverage in a time-resolved fashion. We adjust our model to suit several blood flow patterns and demonstrate the algorithm’s efficiency on two- and three-dimensional simulations.

(The talk will be given in English)

Speaker:     Prof. Shlomo Shamai
                    Distinguished Professor, Chair in Communications, Technion – Israel Institute of Technology

The Viterbi Faculty of Electrical Engineering, Technion-Israel Institute of Technology

Tuesday, July 3^rd, 2018
14:00 - 15:00

Hall 020 (between Wolfson & Labs Builds.), Faculty of Engineering

Cloud Radio Access Networks, Distributed Information Bottleneck, and more: A Unified Information Theoretic View

Abstract

We consider transmission over a cloud radio access network (CRAN) focusing on the framework of oblivious processing at the relay nodes (radio units), i.e., the relays are not cognizant of the users' codebooks.

This approach is motivated by future wireless communications (5G and beyond) and the theoretical results connect to a variety of different information theoretic models and problems.

First it is shown that relaying a-la Cover-El Gamal, i.e., compress-and-forward with joint decompression and decoding, which reflects 'noisy network coding,' is optimal.

The penalty of obliviousness is also demonstrated to be at most a constant gap, when compared to cut-set bounds.

Naturally, due to the oblivious (nomadic) constraint the CRAN problem intimately comments to Chief Executive Officer (CEO) source(s) coding under a logarithmic loss distortion measure.

Furthermore, we identify and elaborate on some interesting connections with the distributed information bottleneck model for which we characterize optimal tradeoffs between rates (i.e., complexity) and information (i.e., accuracy) in the discrete and vector Gaussian frameworks.

Further connections to 'information combining' and 'common reconstruction' are also pointed out. In the concluding outlook, some interesting problems are mentioned such as the characterization of the optimal input distributions under users' power limitations and rate-constrained compression at the relay nodes,

---------------------------------------------------------------------------

Joint work with: I.E. Aguerri (Paris Research Center, Huawei France) A. Zaidi (Universite Paris-Est, Paris) and G. Caire (USC-LA and TUB, Berlin)

The research is supported by the European Union's Horizon 2020 Research And Innovation Programme: no. 694630.

‏

‏

(The talk will be given in English)

Speaker:     Prof. Lihi Zelnik-Manor
                   Faculty of Electrical Engineering, Technion

Monday, June 4^th, 2018
15:00 - 16:00

Room 011, Kitot Bldg., Faculty of Engineering

Maintaining Internal Image Statistics in Synthesized Images
 

Abstract

Recent work has shown impressive success in automatically creating new images with desired properties such as transferring painterly style, modifying facial expressions or manipulating the center of attention of the image. In this talk I will discuss two of the standing challenges in image synthesis and how we tackle them:
- I will describe our efforts in making the synthesized images more photo-realistic.
- I will further show how we can broaden the scope of data that can be used for training synthesis networks, and with that provide a solution to new applications.

BIO

Lihi Zelnik-Manor is an Associate Professor in the Faculty of Electrical Engineering in the Technion, Israel, where she also serves as the Academic Liaison between Technion and CornellTech. Prior to the Technion, she worked as a post-doctoral fellow in the Department of Engineering and Applied Science in the California Institute of Technology (Caltech). She holds a PhD and MSc (with honors) in Computer Science from the Weizmann Institute of Science and a BSc (summa cum laude) in Mechanical Engineering from the Technion. She was a visiting Associate Professor at CornellTech between 2014 and 2016.
Prof. Zelnik-Manor’s awards and honors include the Sloan-Swartz postdoctoral fellowship, the best Student Paper Award at the IEEE SMI'05, the AIM@SHAPE Best Paper Award 2005, and the people's choice award at WACV2018. She is also a recipient of the Gutwirth prize for the promotion of research and several grants from ISF, MOST, the 7th European R&D Program, and others.
Prof Zelnik-Manor has served as Area Chair for ECCV and CVPR multiple times, as Program Chair of CVPR’16 and as Associate Editor at TPAMI. She has further had industrial collaborations with Intel, Adobe, and Microsoft Research.