BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Adaptive and Move Making Auxiliary Cuts for Binary
Pairwise Energies - Olga Veksler (University of W
estern Ontario)
DTSTART;TZID=Europe/London:20170908T120000
DTEND;TZID=Europe/London:20170908T125000
UID:TALK78481AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/78481
DESCRIPTION:Co-author: Lena Gorelick (University of Wes
tern Ontario)
Many com
puter vision problems require optimization of bina
ry non-submodular energies. In this context\, loca
l iterative submodularization techniques based on
trust region (LSA-TR) and auxiliary functions (LSA
-AUX) have been recently proposed. They achieve s
tate-of-the-art-results on a number of computer vi
sion applications. We extend the LSA-AUX framework
in two directions. First\, unlike LSA-AUX\, which
selects auxiliary functions based solely on the c
urrent solution\, we propose to incorporate severa
l additional criteria. This results in tighter bou
nds for configurations that are more likely or clo
ser to the current solution. Second\, we propose m
ove-making extensions of LSA-AUX which achieve tig
hter bounds by restricting the search space. Final
ly\, we evaluate our methods on several applicatio
ns. We show that for each application at least one
of our extensions significantly outperforms the o
riginal LSA-AUX. Moreover\, the best extension of
LSA-AUX is comparable to or better than LSA-TR on
four out of six applications.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR