Event
Ph.D. Dissertation Defense: Saketh Rambhatla
Friday, December 2, 2022
3:00 p.m.
IRB 4105
Emily Irwin
301 405 0680
eirwin@umd.edu
ANNOUNCEMENT: Ph.D. Defense
Name: Saketh Rambhatla
Committee:
Professor Rama Chellappa (Chair)
Professor Abhinav Shrivastava (Co-chair)
Professor Behtash Babadi
Professor Christopher Metzler
Professor Carl Vondrick (Special Member)
Professor Wojciech Czaja (Dean's Representative)
Date/time: Dec 2nd 2022 (12/02/2022) 3-5pm
Location: IRB 4105
Zoom link: https://umd.zoom.us/j/3932406723
Title: Towards in-the-wild visual understanding
Abstract: Computer vision research has seen tremendous success in recent times . This success can be attributed to the recent breakthroughs in deep learning technology and such systems have been shown to achieve superhuman performance on several academic datasets. Contingent on this success, these systems are actively being deployed in several household and industrial appli- cations like robotics. However, current systems perform poorly when deployed in the real world, a.k.a in-the-wild, as most of the assumptions made during the modeling stage are violated. For example, consider object detectors, they require clean data for training, and they have no way of detecting or rejecting novel categories not seen in the data. In this thesis, we systematically identify problems that arise in a typical learning setup, the input, model, and the output, and propose effective solutions to mitigate them.