BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Reconstruction and Applications of Collective Storylines from Web 
 Photo Collections - Gunhee Kim\, Carnegie Mellon University
DTSTART:20130416T120000Z
DTEND:20130416T130000Z
UID:TALK44567@talks.cam.ac.uk
CONTACT:32739
DESCRIPTION:Widespread access to photo-taking devices and high speed Inter
 net has combined with rampant social networking to produce an explosion in
  picture sharing on web platforms. In this environment\, new challenges in
  image acquisition\, processing\, and sharing have emerged\, creating exci
 ting opportunities for research in computer vision and multimedia data min
 ing. I explore several of these interesting problems in my work reconstruc
 ting collective storylines from large-scale\, unordered\, online image col
 lections. \nIn this talk\, I begin by introducing the two fundamental chal
 lenges faced in this research: modelling the temporal trends of whole imag
 e collections and detecting repetitive content among images. I present pro
 posed solutions to each problem\, along with interesting applications enab
 led by the results. First\, I discuss our approach for jointly aligning an
 d segmenting a large number of photo streams from multiple users\, as a fi
 rst technical step to uncovering the common storylines of outdoor recreati
 onal activities such as scuba diving and horse riding. Second\, I present 
 our results for visualizing pictorial stories associated with the competin
 g brands such as sports brands Nike and Adidas. I conclude my presentation
  with a discussion of future projects to further extend my research.
LOCATION:Small Lecture Theatre\, Microsoft Research Ltd\, 21 Station Road\
 , Cambridge\, CB1 2FB
END:VEVENT
END:VCALENDAR