BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Reconstruction meets recognition
DTSTART:20140403T140000
DTEND:20140403T150000
DTSTAMP:20260407T055428Z
UID:a8f7e88276c7bdbf0ed0a486ba7b64fc15c5088488bb407ba45d838e
CATEGORIES:Conferences - Seminars
DESCRIPTION:Professor Marc Pollefeys\, Dept. of Computer Science\, ETH Zur
 ich\nAbstract: The two main goals of computer vision are to recognize the 
 different elements of a scene and to reconstruct their 3D shape and spatia
 l arrangement from images.  In the past the reconstruction and recognitio
 n problems have mostly been approached separately.  In recent years a few
  approaches have been proposed that use recognition to help reconstruction
  or vice-versa.  In this talk we present the first approach that jointly 
 performs 3D scene reconstruction and recognition.  Our approach is formul
 ated as a volumetric multi-class segmentation problem and solved using a c
 onvex relaxation method.  A key element of our approach is that the inter
 face between classes (e.g. building\, ground\, vegetation\, air) have diff
 erent anisotropic smoothness priors.  We will show how our joint approach
  significantly improves the results.\nBio: Marc Pollefeys has been a full 
 professor in the Dept. of Computer Science of ETH Zurich since 2007.  Bef
 ore that he was on the faculty at the University of North Carolina at Chap
 el Hill.  He obtained his PhD from the KU Leuven in Belgium in 1999.  Hi
 s main area of research is computer vision\, but he is also active in robo
 tics\, machine learning and computer graphics.  Dr. Pollefeys has receive
 d several prizes for his research\, including a Marr prize\, an NSF CAREER
  award\, a Packard Fellowship and a European Research Council Starting Gra
 nt. He is the author or co-author of more than 200 peer-reviewed publicati
 ons.  He is the general chair of ECCV 2014 in Zurich.
LOCATION:CM 1 105 http://plan.epfl.ch/?lang=fr&room=CM+1105
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR