BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:3D Shape Inference from Images using Deep Learning - Andrew Zisser
 man (University of Oxford)
DTSTART:20171211T160000Z
DTEND:20171211T170000Z
UID:TALK96427@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:The talk will cover two approaches to obtaining 3D shape from 
 images.  First\, we introduce a deep Convolutional Neural Network (ConvNet
 ) architecture that can generate depth maps given a single or multiple ima
 ges of an object. The ConvNet is trained using a prediction loss on both t
 he depth map and the silhouette.  Using a set of sculptures as our 3D obje
 cts\, we show that the ConvNet is able to generalize to new objects\, unse
 en during training\, and that its performance improves given more input vi
 ews of the object.  This is joint work with Olivia Wiles.  Second\, we  us
 e ConvNets to infer 3D shape attributes\, such as planarity\, symmetry and
  occupied space\, from a single image. For this we have assembled an annot
 ated dataset of 150K images of over 2000 different sculptures.  We show th
 at 3D attributes can be learnt from these images and generalize to images 
 of other (non-sculpture) object classes. This is joint work with Abhinav G
 upta and David Fouhey. 
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR