BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Predicting 3D Volume and Depth from a Single View  - Gabriel Brost
 ow (University College London)
DTSTART:20170405T093000Z
DTEND:20170405T103000Z
UID:TALK71903@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:A single glimpse is hardly enough to triangulate the 3D shapes
  of a scene. However\, training examples are readily available\, so statis
 tical models can be trained to map appearance to shape. The details matter
 \, because 3D shapes have different representations and can have many degr
 ees of freedom\, and training data is rarely as clean as we'd wish.\n\nI w
 ill present two separate learning based methods for shape reconstruction\,
  developed by my team at UCL. In the first\, we propose an algorithm that 
 can complete the unobserved geometry of tabletop-sized objects from a sing
 le depth-image. This approach is based on a supervised model trained on al
 ready available volumetric elements. In the second\, instead of a depth-im
 age as input we have just an RGB image\, from which we predict a depth ima
 ge. This is a CNN based method that exploits epipolar geometry constraints
  to learn depth-prediction from binocular pairs\, to overcome the absence 
 of good ground truth depth data. The two systems are not joined\, because 
 there is still more exciting work to be done!\n
LOCATION:Auditorium\, Microsoft Research Ltd\, 21 Station Road\, Cambridge
 \, CB1 2FB
END:VEVENT
END:VCALENDAR