BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:DeepBedMap: A Super-Resolution Generative Adversarial Network for 
 resolving the bed topography of Antarctica - Wei Ji Leong | Antarctic Rese
 arch Centre\, Victoria University
DTSTART:20201020T090000Z
DTEND:20201020T100000Z
UID:TALK152497@talks.cam.ac.uk
CONTACT:Tudor Suciu
DESCRIPTION:A machine learning technique similar to the one used to enhanc
 e everyday photographs is applied to the problem of getting a better pictu
 re of Antarctica's bed - that which is hidden beneath the ice. By taking h
 ints from what satellites can observe at the ice surface\, and training th
 e model on high-resolution ground-truth data\, the novel method is able to
  generate an image of the bed with better topographic characteristics than
  ordinary interpolation methods. The DeepBedMap model is based on an adapt
 ed Enhanced Super-Resolutionv Generative Adversarial Network architecture\
 , chosen to minimize per-pixel elevation errors while producing realistic 
 topography. The final product is a four times upsampled (250 m) bed elevat
 ion model of Antarctica that can be used by scientists running fine scale 
 ice sheet models relevant for predicting future sea level trends.\n\n"Zoom
  link":https://us02web.zoom.us/j/89975312802?pwd=dkpvc0M3RGhtY1JOUm5Hc1dCN
 m9IZz09 \n\nMeeting ID: 899 7531 2802 \n\nPasscode: 945323
LOCATION:https://us02web.zoom.us/j/89975312802?pwd=dkpvc0M3RGhtY1JOUm5Hc1d
 CNm9IZz09
END:VEVENT
END:VCALENDAR