BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Probabilistic Systems\, Information\, and Inferenc
 e Group Seminars
SUMMARY:Tracking intentionality using behavioural models a
 nd Bayesian methods.  - Prof. Simon Godsill\, CUED
DTSTART;TZID=Europe/London:20161020T150000
DTEND;TZID=Europe/London:20161020T160000
UID:TALK68624AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/68624
DESCRIPTION:In this talk I will describe recent methods and ap
 plications for high-level inference and tracking o
 f multiple object\, groups and networks\, by incor
 poration of behavioural interactions and unobserve
 d intent into the dynamical models. The idea here 
 is to extend the standard multiple object tracking
  paradigm to one in which we may automatically lea
 rn dynamic interactions between those objects\, as
  well as infer possible intentionalities of the ob
 jects. Our models are based on principles from ani
 mal behavioural analysis in which objects follow p
 atterns of behaviour based loosely upon what their
  neighbours in the group are doing\, and upon the 
 (unknown) intentionality of the group\, for exampl
 e its final destination. We may also learn more co
 mplex interactions such as whether one member of t
 he group is a `leader' of the dynamics and how the
  objects are split between different groupings. Mo
 dels are typically formulated in continuous time\,
  and inference is carried out on-line using numeri
 cal Bayesian filtering strategies\, implemented wi
 th state of the art methods such as particle filte
 rs and Markov chain Monte Carlo. Applications will
  be presented from the areas of vehicle tracking\,
  wild animal pack hunting behaviour analysis\, fin
 ancial time series\, and finally applications in U
 ser Interfaces for automobiles in which the task i
 s to determine accurately and rapidly the intended
  icon a user is pointing at on a screen\, based on
  the trajectory of hand motion near to the screen\
 , and in the presence of disturbances from suspens
 ion and road surface.
LOCATION:LR4\, Cambridge University Engineering Department
CONTACT:Prof. Ramji Venkataramanan
END:VEVENT
END:VCALENDAR