A configurable architecture for a visual saliency system and its application in retail

Nandhini Chandramoorthy, Siddarth Advani, Kevin M. Irick, Vijaykrishnan Narayanan

Research output: Contribution to conferencePaperpeer-review

Abstract

Visual attention in a scene has been the focus of study for many years and several theories have been advanced to explain human attention behaviour and fixation patterns. Of particular interest to us is the concept of bottom-up visual stimulus-driven saliency. AIM (Attention based on Information Maximization) is a computational model that mimics the interaction between a local cell's activation and the activation of cells in the larger surrounding neighborhood. AIM explains the concept of visual salience in terms of the amount of information present in an image patch within the context of its surround - i.e. how likely the patch is to appear with respect to its surround.

Original languageEnglish (US)
Number of pages1
DOIs
StatePublished - Aug 12 2013
Event21st Annual International IEEE Symposium on Field-Programmable Custom Computing Machines, FCCM 2013 - Seattle, WA, United States
Duration: Apr 28 2013Apr 30 2013

Other

Other21st Annual International IEEE Symposium on Field-Programmable Custom Computing Machines, FCCM 2013
CountryUnited States
CitySeattle, WA
Period4/28/134/30/13

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Software
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'A configurable architecture for a visual saliency system and its application in retail'. Together they form a unique fingerprint.

Cite this