BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Monocular 3D Pose Estimation - Srimal Jayawardena\, PhD Candidate\
 , Research School of Computer Science at the Australian National Universit
 y (ANU)
DTSTART:20111122T141500Z
DTEND:20111122T150000Z
UID:TALK33691@talks.cam.ac.uk
CONTACT:Rachel Fogg
DESCRIPTION:The problem of identifying the 3D pose of a known object from 
 a given 2D image has important applications in Computer Vision. Our propos
 ed method of registering a 3D model of a known object on a given 2D photo 
 of the object has numerous advantages over existing methods. It does not r
 equire prior training\, knowledge of the camera parameters\, explicit poin
 t correspondences or matching features between the image and model. Unlike
  techniques that estimate a partial 3D pose (as in an overhead view of tra
 f?c or machine parts on a conveyor belt)\, our method estimates the comple
 te 3D pose of the object. It works on a single static image from a given v
 iew under varying and unknown lighting conditions. For this purpose we der
 ive a novel illumination-invariant distance measure between the 2D photo a
 nd projected 3D model\, which is then minimised to find the best pose para
 meters. Results for vehicle pose detection in real photographs are present
 ed.
LOCATION: Room BN4-38\, Cambridge University Engineering Department
END:VEVENT
END:VCALENDAR