BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning to see with deep learning architectures for localisation 
 and scene understanding - Alex Kendall (University of Cambridge)
DTSTART:20160526T101500Z
DTEND:20160526T111500Z
UID:TALK66113@talks.cam.ac.uk
CONTACT:Mariana Marasoiu
DESCRIPTION:*Abstract*\n\nWe can now teach machines to recognize objects. 
 However\, in order to teach a machine to “see” we need to understand g
 eometry as well as semantics. Given an image of a road scene\, for example
 \, an autonomous vehicle needs to determine where it is\, what's around it
 \, and what's going to happen next? This requires not only object recognit
 ion\, but depth\, motion and spatial perception\, and instance-level ident
 ification. We present work towards solving these problems using deep learn
 ing.\n\nThe first\, SegNet\, is a deep convolutional network architecture 
 designed to map input RGB images to pixel labels for scene understanding. 
 It is composed of an encoder network and a decoder network which ends with
  a soft­max classifier. The entire architecture can be trained end-to-end
  using stochastic gradient descent. SegNet can produce a dense pixel-wise 
 output in real-time with a measure of model uncertainty. We show SegNet ap
 plied to both classification and regression tasks.\n\nSecondly\, PoseNet i
 s a real-time relocalisation system. We show how to train very deep networ
 ks to regress the camera's 3D position and orientation from a single image
 . The algorithm can operate over large scale indoor and outdoor areas in r
 eal time.\n\nLive web demonstrations and links to publications can be foun
 d on our project webpages:\nhttp://mi.eng.cam.ac.uk/projects/segnet/\nhttp
 ://mi.eng.cam.ac.uk/projects/relocalisation/
LOCATION:SS03 Meeting Room\, Computer Laboratory
END:VEVENT
END:VCALENDAR