File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Nimble: A kernel density model of saccade-based visual memory

TitleNimble: A kernel density model of saccade-based visual memory
Authors
KeywordsComputational Modeling
Eye Movements
Face Recognition
Kernel Density Estimation
Memory
Object Recognition
Issue Date2008
PublisherAssociation for Research in Vision and Ophthalmology. The Journal's web site is located at http://wwwjournalofvisionorg/
Citation
Journal Of Vision, 2008, v. 8 n. 14, article 17 How to Cite?
AbstractWe present a Bayesian version of J. Lacroix, J. Murre, and E.. Postma's- (2006) Natural Input Memory (NIM) model of saccadic visual memory. Our model, which we call NIMBLE (NIM with Bayesian Likelihood Estimation), uses a cognitively plausible image sampling technique that provides a foveated representation of image patches. We conceive of these memorized image fragments as samples from image class distributions and model the memory of these fragments using kernel density estimation. Using these models, we derive class-conditional probabilities of new image fragments and combine individual fragment probabilities to classify images. Our Bayesian formulation of the model extends easily to handle multi-class problems. We validate our model by demonstrating human levels of performance on a face recognition memory task and high accuracy on multi-category face and object identification. We also use NIMBLE to examine the change in beliefs as more fixations are taken from an image. Using fixation data collected from human subjects, we directly compare the performance of NIMBLE's memory component to human performance, demonstrating that using human fixation locations allows NIMBLE to recognize familiar faces with only a single fixation. © ARVO.
Persistent Identifierhttp://hdl.handle.net/10722/169054
ISSN
2015 Impact Factor: 2.341
2015 SCImago Journal Rankings: 1.042
ISI Accession Number ID
References

 

DC FieldValueLanguage
dc.contributor.authorBarrington, Len_US
dc.contributor.authorMarks, TKen_US
dc.contributor.authorHsiao, JHWen_US
dc.contributor.authorCottrell, GWen_US
dc.date.accessioned2012-10-08T03:41:14Z-
dc.date.available2012-10-08T03:41:14Z-
dc.date.issued2008en_US
dc.identifier.citationJournal Of Vision, 2008, v. 8 n. 14, article 17en_US
dc.identifier.issn1534-7362en_US
dc.identifier.urihttp://hdl.handle.net/10722/169054-
dc.description.abstractWe present a Bayesian version of J. Lacroix, J. Murre, and E.. Postma's- (2006) Natural Input Memory (NIM) model of saccadic visual memory. Our model, which we call NIMBLE (NIM with Bayesian Likelihood Estimation), uses a cognitively plausible image sampling technique that provides a foveated representation of image patches. We conceive of these memorized image fragments as samples from image class distributions and model the memory of these fragments using kernel density estimation. Using these models, we derive class-conditional probabilities of new image fragments and combine individual fragment probabilities to classify images. Our Bayesian formulation of the model extends easily to handle multi-class problems. We validate our model by demonstrating human levels of performance on a face recognition memory task and high accuracy on multi-category face and object identification. We also use NIMBLE to examine the change in beliefs as more fixations are taken from an image. Using fixation data collected from human subjects, we directly compare the performance of NIMBLE's memory component to human performance, demonstrating that using human fixation locations allows NIMBLE to recognize familiar faces with only a single fixation. © ARVO.en_US
dc.languageengen_US
dc.publisherAssociation for Research in Vision and Ophthalmology. The Journal's web site is located at http://wwwjournalofvisionorg/en_US
dc.relation.ispartofJournal of Visionen_US
dc.subjectComputational Modelingen_US
dc.subjectEye Movementsen_US
dc.subjectFace Recognitionen_US
dc.subjectKernel Density Estimationen_US
dc.subjectMemoryen_US
dc.subjectObject Recognitionen_US
dc.titleNimble: A kernel density model of saccade-based visual memoryen_US
dc.typeArticleen_US
dc.identifier.emailHsiao, JHW:jhsiao@hku.hken_US
dc.identifier.authorityHsiao, JHW=rp00632en_US
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.doi10.1167/8.14.17en_US
dc.identifier.scopuseid_2-s2.0-56349100680en_US
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-56349100680&selection=ref&src=s&origin=recordpageen_US
dc.identifier.volume8en_US
dc.identifier.issue14en_US
dc.identifier.spagearticle 17-
dc.identifier.epagearticle 17-
dc.identifier.isiWOS:000262231300018-
dc.publisher.placeUnited Statesen_US
dc.identifier.scopusauthoridBarrington, L=14041197300en_US
dc.identifier.scopusauthoridMarks, TK=24780830500en_US
dc.identifier.scopusauthoridHsiao, JHW=7101605473en_US
dc.identifier.scopusauthoridCottrell, GW=7102792906en_US

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats