BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Beyond Mindless Labeling: *Really* Leveraging Humans to Build Inte
 lligent Machines  - Devi Parikh\, Virginia Tech
DTSTART:20140904T100000Z
DTEND:20140904T110000Z
UID:TALK54111@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Human ability to understand natural images far exceeds machine
 s today. One reason for this gap is an artificially restrictive learning s
 et up – humans today teach machines via Morse code (e.g. providing binar
 y labels on images\, such as “this is a horse” or “this is not”)\,
  and machines are typically silent. These systems have the potential to be
  significantly more accurate if they tap into the vast common-sense knowle
 dge humans have about the visual world. I will talk about our work on enri
 ching the communication between humans and machines by exploiting mid-leve
 l visual properties or attributes. I will also talk about the more difficu
 lt problem of directly learning common-sense knowledge simply by observing
  the structure of the visual world around us. Unfortunately\, this require
 s automatic and accurate detection of objects\, their attributes\, poses\,
  etc. in images\, leading to a chicken-and-egg problem. I will argue that 
 the solution is to give up on photorealism. Specifically\, I will talk abo
 ut our work on exploiting human-generated abstract visual scenes to learn 
 common-sense knowledge and study high-level vision problems. 
LOCATION:Auditorium\, Microsoft Research Ltd\, 21 Station Road\, Cambridge
 \, CB1 2FB
END:VEVENT
END:VCALENDAR