BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Images as Sets of Locally Weighted Features - Teo de Campos\, Univ
 ersity of Surrey
DTSTART:20100601T130000Z
DTEND:20100601T140000Z
UID:TALK24959@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:*Abstract:* I`ll present a novel image representation in which
  images are modeled as order-less sets of weighted visual features. Each v
 isual feature is associated with a weight factor that may inform its relev
 ance. This framework can be applied to various bag-of-patches approaches s
 uch as the bag-of-visual-word or the Fisher kernel representations. We sug
 gest that if dense sampling is used\, different schemes to weight local fe
 atures can be evaluated\, leading to results that are often better than th
 e combination of multiple sampling schemes\, at a much lower computational
  cost\, because the features are extracted only once. This allows our fram
 ework to be a testbed for saliency estimation methods in image categorizat
 ion tasks. We explored two main possibilities for the estimation of local 
 feature relevance. The first one is based on the use of saliency maps obta
 ined from human feedback\, either by gaze tracking or by mouse clicks. The
  method is able to profit from such maps\, leading to a significant improv
 ement in categorization performance. The second possibility is based on au
 tomatic saliency estimation methods\, including Itti&Koch`s method and the
  SIFT`s DoG. We evaluated the proposed framework and saliency estimation m
 ethods using an in house dataset and the PASCAL VOC 2008/2007 dataset\, sh
 owing that some of the saliency estimation methods lead to a significant p
 erformance in comparison to the standard unweighted representation. \n
LOCATION:Small public lecture room\, Microsoft Research Ltd\, 7 J J Thomso
 n Avenue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR