BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium : A variety of manifold methods in Computer Vision
DTSTART:20150922T161500
DTEND:20150922T173000
DTSTAMP:20260428T114536Z
UID:8e7c569decb2381c76b8f93ab30e1b861f5d08372b3716940fbf137b
CATEGORIES:Conferences - Seminars
DESCRIPTION:By : Richard Hartley - Australian National UniversityAbstract 
 :\nI will talk about uses of Riemannian manifolds in computer vision\, Exa
 mples of manifolds in CV include Kendall's Shape manifold\, Grassman manif
 olds (used in recognition) the manifold of rotations\, SO3\, and positive-
 definite matrices.  I will focus on two topics: kernel-learning methods o
 n manifolds and optimization on manifolds.\nThe identification of kernels 
 on manifolds allows us to apply techniques such as kernel-SVM and kernel d
 ictionary-learning on spaces such as the Grassman manifold\, with excellen
 t results. Optimization methods directly on manifolds allow us to approach
  different geometric problems through techniques such as rotation averagin
 g with different robust cost functions (Huber\, L1)Bio :\nRichard Hartley 
 is a member of the computer vision group in the Research School of Enginee
 ring\, at the Australian National University\, where he has been since Jan
 uary\, 2001. He is a joint leader of the Computer Vision group in NICTA\, 
 a government funded research laboratory.\nDr. Hartley worked at the Genera
 l Electric Research and Development Center from 1985 to 2001\, working fir
 st in VLSI design\, and later in computer vision.  He became involved wit
 h Image Understanding and Scene Reconstruction working with GE's Simulatio
 n and Control Systems Division.\nIn 1991\, he began an extended research e
 ffort in the area of applying projective geometry techniques to reconstruc
 tion using calibrated and semi-calibrated cameras. This research direction
  was one of the dominant themes in computer vision research throughout the
  1990s. In 2000\, he co-authored (with Andrew Zisserman) a book on Multivi
 ew Geometry in Computer Vision\, summarizing the previous decade’s resea
 rch in this area.More information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR