BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:3D Reconstruction using Point-Based Fusion - Damien Lefloch\, Seig
 en University Germany
DTSTART:20130821T133000Z
DTEND:20130821T140000Z
UID:TALK46826@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Since real-time range cameras (Time-of-Flight or Kinect camera
 s) become more and more available in the market\, new 3D applications aris
 e such as scene understanding\, augmented reality or human-computer intera
 ction. Recently\, a cheap 3D reconstruction method using the kinect camera
  has been proposed. But its extension for ToF cameras has not yet been dem
 onstrated and will bring further challenges due to ToF camera properties (
 e.g. low resolution\, complex noise behaviour\, etc.).\n\nThis research fo
 cuses on developing a fast 3D geometry reconstruction based on a simple po
 ints representation that allows adaptive resolution and workspace. Our met
 hod uses the iterative closest point (ICP) to track the camera motion (des
 cribing all data in the same reference)\, fuse all similar data to a singl
 e vertex list (model) and visualizes this model using surfel rendering. A 
 surfel is composed of a 3D position (center)\, an orientation (normal) and
  a size (radius). Note that the data fusion module assumes a simple gaussi
 an distribution.\n\nSince no volumetric data is used\, our method can be e
 asily extended to dynamics environment. Dynamic objects are detected using
  the matching measurement provided by the ICP algorithm. A GPU based regio
 n growing method is later applied to the outlier map and the input data wi
 th respect to depth and normal similarities in order to fully segment dyna
 mic objects.\n\nThe project was a collaboration with Microsoft Research Ca
 mbridge and University College London.\n
LOCATION:Auditorium\, Microsoft Research Ltd\, 21 Station Road\, Cambridge
 \, CB1 2FB
END:VEVENT
END:VCALENDAR