BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Unsupervised Iceberg Detection in Satellite Images Using Recurrent
  Principal Component Analysis - Anita Faul | British Antarctic Survey
DTSTART:20201013T100000Z
DTEND:20201013T113000Z
UID:TALK152494@talks.cam.ac.uk
CONTACT:Tudor Suciu
DESCRIPTION:The shape and distribution of icebergs and their trajectory th
 rough the oceans from source\, to where they break up and melt\, can be di
 agnostic of ice-sheet dynamics\, and ocean and atmospheric conditions. Exi
 sting methods to detect icebergs in satellite imagery are optimised for op
 en ocean and work less well for icebergs within the sea ice pack. Differen
 t conditions for each satellite image also mean that one method does not f
 it all situations. This talk explores the possibility of making the algori
 thm data dependent while touching upon specific difficulties in the develo
 pment of the algorithm which are intrinsic to other data science problems.
 I will also ponder the value of simplicity.
LOCATION:https://ukri.zoom.us/j/91046719352
END:VEVENT
END:VCALENDAR