BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Vienna
BEGIN:DAYLIGHT
DTSTART:20180325T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20171029T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260424T143114Z
UID:5a7c07876b13b077582501@ist.ac.at
DTSTART:20180226T160000
DTEND:20180226T170000
DESCRIPTION:Speaker: Alexander Kolesnikov\nAbstract: In this thesis we focu
 s on the alternative way of tackling the aforementioned issue.We concentra
 te on methods\, which make use of weakly-labeled or even unlabeled data. S
 pecifically\, the first half of the thesis is dedicated to the semantic im
 age segmentation task. We develop a technique\, which achieves competitive
  segmentation performance and only requires annotations in a form of globa
 l image-level labels instead of dense segmentation masks. Subsequently\, w
 e present a new methodology\, which further improves segmentation performa
 nce by leveraging tiny additional feedback from a human annotator. By usin
 g our methods practitioners can greatly reduce the amount of data annotati
 on effort\, which is required to learn modern image segmentation models.In
  the second half of the thesis we focus on methods for learning from unlab
 eled visual data. We study a family of autoregressive models for modeling 
 structure of natural images and discuss potential applications of these mo
 dels. Moreover\, we conduct in-depth study of one of these applications\, 
 where we develop the state-of-the-art model for the probabilistic image co
 lorization task.
LOCATION:Mondi Seminar Room 2\, Central Building\, ISTA
ORGANIZER:useiss@ist.ac.at
SUMMARY:Alexander Kolesnikov: Alex Kolesnikov (Lampert group): Weakly-Super
 vised Segmentation and Unsupervised Modeling of Natural Images
URL:https://talks-calendar.ista.ac.at/events/1105
END:VEVENT
END:VCALENDAR