October 27, 2020

12:00 pm / 1:15 pm

Venue

ZOOM

Rama Chellappa
Johns Hopkins

Recorded Seminar: https://wse.zoom.us/rec/share/eyf1_jff6ce8-Xw6wmsY7l-rJwYnn-VmC5i-lUYwHv3bv5hFS23HGCLQmrIyync.pO__BFBQf8ym7V8S?startTime=1603813572000

?Generationsof Generative Models for Images and Videoswith Applications?

Abstract:  We will provide a brief historicalperspective on time series models, MRFs, Gibbs distributions, fractals, mosaicmodels, dynamic texture models, discuss their advantages and drawbacks and thenmove onto a detailed exposition of GANs. Will show applications of GANs forimage synthesis,  image restoration, medical image reconstruction, domainadaptation and defending against adversarial attacks.

Bio:  Prof.Rama Chellappa is a BloombergDistinguished Professor in the Departments ofElectrical and ComputerEngineering and Biomedical Engineering with a secondary appointment in theDepartment of Computer Science at Johns Hopkins University (JHU). At JHU, he isalso affiliated with CIS, CLSP, Malone Centerand MINDS. Before coming to JHUin August 2020, he was a Distinguished University Professor, a MintaMartin Professor of Engineering, and a Professor in the ECE department and theUniversity of Maryland Institute Advanced Computer Studies at the University ofMaryland (UMD).  He holds a non-tenureposition as a College Park Professor in the ECE department at UMD. His currentresearcher interests are computer vision, pattern recognition, machineintelligence and artificial intelligence. He has received numerous research,teaching, service, and mentoring awards from the University of SouthernCalifornia, University of Maryland, IBM, IEEE Computer and Signal ProcessingSocieties, the IEEE Biometrics Council and the International Association ofPattern Recognition. He is a Fellow of IEEE, IAPR, OSA, AAAS, ACM, and AAAI andholds six patents.