BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Robust Surface Reconstruction
DTSTART:20150527T140000
DTSTAMP:20260429T004625Z
UID:08581a0170827b23b769cbf826cdfe5ee72a429769c6c5b0ce6430b4
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr Virginia Estellers\, UCLA\nWe propose a method to reconstru
 ct surfaces from oriented point clouds with non-uniform sampling and noise
  by formulating the problem as a convex minimization that reconstructs the
  indicator function of the surface’s interior. Compared to previous mode
 ls\, our reconstruction is robust to noise and outliers because it substit
 utes the least-squares fidelity term by a robust Huber penalty\; this allo
 ws to recover sharp corners and avoids the shrinking bias of least squares
 . We choose an implicit parametrization to reconstruct surfaces of unknown
  topology and close large gaps in the point cloud.\nFor an efficient repre
 sentation\, we approximate the implicit function by a hierarchy of locally
  supported basis elements adapted to the geometry of the surface. Unlike a
 d-hoc bases over an octree\, our hierarchical B-splines from isogeometric 
 analysis locally adapt the mesh and degree of the splines during reconstru
 ction. The hierarchical structure of the basis speeds-up the minimization 
 and efficiently represents clustered data. We also advocate for convex opt
 imization\, instead isogeometric finite-element techniques\, to efficientl
 y solve the minimization and allow for non-differentiable functionals. Exp
 eriments show state-of-the-art performance within a more flexible framewor
 k. This is joint work with Michael Scott and Stefano Soatto.\nBio: Virgini
 a Estellers received her PhD in image processing from Ecole Polythechnique
  Federale de Lausanne in 2013\, and joined the UCLA Vision Lab as a postdo
 ctoral fellow with an SNSF fellowship. Previous to that\, she did her unde
 rgraduate studies at the Polytechnic University of Catalonia in both Mathe
 matics and Electrical Engineering\nHer research interests are in mathemati
 cal modeling and computational techniques in imaging and vision\, particul
 arly using variational methods\, convex optimization\, and partial differe
 ntial equations. She is interested in the theoretical and physical aspects
  of the acquisition of images\, their mathematical representations\, and t
 he development of efficient algorithms to extract information from them.
LOCATION:ELG120 http://plan.epfl.ch/?lang=en&room=elg120
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR