Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Searching and Browsing Social Images through iAVATAR Towards Yahoo! Challenge 2010: Novel Image Understanding
 

Summary: Searching and Browsing Social Images through iAVATAR
Towards Yahoo! Challenge 2010: Novel Image Understanding
Aixin Sun Sourav S. Bhowmick
School of Computer Engineering, Nanyang Technological University, Singapore
axsun|assourav@ntu.edu.sg
ABSTRACT
We present a novel graphical social images retrieval system, called
iAVATAR (interActive VisuAl-representative TAgs Relationship),
that takes a concrete step to address social image search and under-
standing challenge. iAVATAR is the first systematic effort to explore
relationships between tags by visual-representativeness, frequency,
and co-occurrence, coherently. Through the visualization of tag
properties and relationships, iAVATAR offers users a visual expla-
nation of the retrieved social images.
1. OVERVIEW
Availability of social tags has significantly facilitated web im-
age search and organization as this rich collection of tags provides
more information than we can possibly extract from the content-
based algorithms. However, it has been widely recognized that re-
alizing a tag-based image retrieval system is technically challeng-

  

Source: Aixin, Sun - School of Computer Engineering, Nanyang Technological University

 

Collections: Computer Technologies and Information Sciences