University of Cambridge > Talks.cam > Cambridge Centre for Analysis talks > Natural user interface and machine perception of gestures

Natural user interface and machine perception of gestures

Add to your list(s) Download to your calendar using vCal

  • UserAndrew Blake, Microsoft Research Cambridge
  • ClockThursday 24 November 2011, 16:00-17:00
  • HouseMR9, CMS.

If you have a question about this talk, please contact CCA.

CCA Industrial Seminar

The problem of programming a computer to recognise the motion of the human body has been studied over two decades or so. In common with other problems in machine perception, variability (across the class of human bodies and their possible movements) is best addressed by computing with probabilities. A number of approaches to building suitable probabilistic models has been considered, for example Monte Carlo Inference and hierarchically structured models. Most recently the problem has been solved at commercial scale in Microsoft’s “Kinect” system, using an active stereo camera which has entered the Guinness book of records as the fastest-selling piece of electronic hardware ever made. The approach to perception adopted in Kinect, which is perhaps a little surprising, will be described in the lecture.

Speaker: Andrew Blake

Bio: http://research.microsoft.com/~ablake

This talk is part of the Cambridge Centre for Analysis talks series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2020 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity