BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260610T160216EDT-2197WooVWh@132.216.98.100
DTSTAMP:20260610T200216Z
DESCRIPTION:Spectral Matching & Learning of Surface Data - Example on Brain
  Surfaces.\n\nHow to analyze complex shapes\, such as of the highly folded
  surface of the brain?  In this talk\, I will show how spectral representa
 tions of shapes can benefit neuroimaging and\, more generally\, problems w
 here data fundamentally lives on surfaces.  Key operations\, such as segme
 ntation and registration\, typically need a common mapping of surfaces\, o
 ften obtained via slow and complex mesh deformations in a Euclidean space.
   Here\, we exploit spectral coordinates derived from the Laplacian eigenf
 unctions of shapes and also address the inherent instability of spectral s
 hape decompositions.  Spectral coordinates have the advantage over Euclide
 an coordinates\, to be geometry aware and to parameterize surfaces explici
 tly.  This change of paradigm\, from Euclidean to spectral representations
 \, enables a classifier to be applied *directly* on surface data\, via spe
 ctral coordinates.   The talk will focus\, first\, on spectral representat
 ions of shapes\, with an example on brain surface matching\, and second\, 
 on the learning of surface data\, with an example on automatic brain surfa
 ce parcellation.   \n
DTSTART:20171113T190000Z
DTEND:20171113T200000Z
LOCATION:Room 5340\, CA\, Pav. André-Aisenstadt
SUMMARY:Hervé Lombaert\, ETS
URL:https://www.mcgill.ca/mathstat/channels/event/herve-lombaert-ets-282660
END:VEVENT
END:VCALENDAR