Robust visual estimation as source separation

Mordechai Z. Juni, Manish Singh, Laurence T. Maloney

Research output: Contribution to journalArticle

Abstract

We developed a method analogous to classification images that allowed us to measure the influence that each dot in a dot cluster had on observers' estimates of the center of the cluster. In Experiment 1, we investigated whether observers employ a robust estimator when estimating the centers of dot clusters that were drawn from a single distribution. Most observers' fitted influences did not differ significantly from that predicted by a center-of-gravity (COG) estimator. Such an estimator is not robust. In Experiments 2 and 3, we considered an alternative approach to the problem of robust estimation, based on source separation, that makes use of the visual system's ability to segment visual data. The observers' task was to estimate the center of one distribution when viewing complex dot clusters that were drawn from a mixture of two distributions. We compared human performance to that of an ideal observer that separated the cluster into two sources through a maximum likelihood algorithm and based its estimates of location using the dots it assigned to just one of the two sources. The results suggest that robust methods employed by the visual system are closely tied to mechanisms of perceptual segmentation.

Original languageEnglish (US)
Pages (from-to)1-20
Number of pages20
JournalJournal of Vision
Volume10
Issue number14
DOIs
StatePublished - 2010

Fingerprint

Gravitation

Keywords

  • Center of gravity
  • Perceptual segmentation
  • Robust statistics
  • Scission
  • Source separation

ASJC Scopus subject areas

  • Ophthalmology
  • Sensory Systems

Cite this

Robust visual estimation as source separation. / Juni, Mordechai Z.; Singh, Manish; Maloney, Laurence T.

In: Journal of Vision, Vol. 10, No. 14, 2010, p. 1-20.

Research output: Contribution to journalArticle

Juni, Mordechai Z. ; Singh, Manish ; Maloney, Laurence T. / Robust visual estimation as source separation. In: Journal of Vision. 2010 ; Vol. 10, No. 14. pp. 1-20.
@article{828cd198789a4401acdf665e42910b7d,
title = "Robust visual estimation as source separation",
abstract = "We developed a method analogous to classification images that allowed us to measure the influence that each dot in a dot cluster had on observers' estimates of the center of the cluster. In Experiment 1, we investigated whether observers employ a robust estimator when estimating the centers of dot clusters that were drawn from a single distribution. Most observers' fitted influences did not differ significantly from that predicted by a center-of-gravity (COG) estimator. Such an estimator is not robust. In Experiments 2 and 3, we considered an alternative approach to the problem of robust estimation, based on source separation, that makes use of the visual system's ability to segment visual data. The observers' task was to estimate the center of one distribution when viewing complex dot clusters that were drawn from a mixture of two distributions. We compared human performance to that of an ideal observer that separated the cluster into two sources through a maximum likelihood algorithm and based its estimates of location using the dots it assigned to just one of the two sources. The results suggest that robust methods employed by the visual system are closely tied to mechanisms of perceptual segmentation.",
keywords = "Center of gravity, Perceptual segmentation, Robust statistics, Scission, Source separation",
author = "Juni, {Mordechai Z.} and Manish Singh and Maloney, {Laurence T.}",
year = "2010",
doi = "10.1167/10.14.1",
language = "English (US)",
volume = "10",
pages = "1--20",
journal = "Journal of Vision",
issn = "1534-7362",
publisher = "Association for Research in Vision and Ophthalmology Inc.",
number = "14",

}

TY - JOUR

T1 - Robust visual estimation as source separation

AU - Juni, Mordechai Z.

AU - Singh, Manish

AU - Maloney, Laurence T.

PY - 2010

Y1 - 2010

N2 - We developed a method analogous to classification images that allowed us to measure the influence that each dot in a dot cluster had on observers' estimates of the center of the cluster. In Experiment 1, we investigated whether observers employ a robust estimator when estimating the centers of dot clusters that were drawn from a single distribution. Most observers' fitted influences did not differ significantly from that predicted by a center-of-gravity (COG) estimator. Such an estimator is not robust. In Experiments 2 and 3, we considered an alternative approach to the problem of robust estimation, based on source separation, that makes use of the visual system's ability to segment visual data. The observers' task was to estimate the center of one distribution when viewing complex dot clusters that were drawn from a mixture of two distributions. We compared human performance to that of an ideal observer that separated the cluster into two sources through a maximum likelihood algorithm and based its estimates of location using the dots it assigned to just one of the two sources. The results suggest that robust methods employed by the visual system are closely tied to mechanisms of perceptual segmentation.

AB - We developed a method analogous to classification images that allowed us to measure the influence that each dot in a dot cluster had on observers' estimates of the center of the cluster. In Experiment 1, we investigated whether observers employ a robust estimator when estimating the centers of dot clusters that were drawn from a single distribution. Most observers' fitted influences did not differ significantly from that predicted by a center-of-gravity (COG) estimator. Such an estimator is not robust. In Experiments 2 and 3, we considered an alternative approach to the problem of robust estimation, based on source separation, that makes use of the visual system's ability to segment visual data. The observers' task was to estimate the center of one distribution when viewing complex dot clusters that were drawn from a mixture of two distributions. We compared human performance to that of an ideal observer that separated the cluster into two sources through a maximum likelihood algorithm and based its estimates of location using the dots it assigned to just one of the two sources. The results suggest that robust methods employed by the visual system are closely tied to mechanisms of perceptual segmentation.

KW - Center of gravity

KW - Perceptual segmentation

KW - Robust statistics

KW - Scission

KW - Source separation

UR - http://www.scopus.com/inward/record.url?scp=79251549811&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79251549811&partnerID=8YFLogxK

U2 - 10.1167/10.14.1

DO - 10.1167/10.14.1

M3 - Article

C2 - 21131562

AN - SCOPUS:79251549811

VL - 10

SP - 1

EP - 20

JO - Journal of Vision

JF - Journal of Vision

SN - 1534-7362

IS - 14

ER -