A saliency-based bottom-up visual attention model for dynamic scenes analysis

David F. Ramirez-Moreno, Odelia Schwartz, Juan F. Ramirez-Villegas

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

This work proposes a model of visual bottom-up attention for dynamic scene analysis. Our work adds motion saliency calculations to a neural network model with realistic temporal dynamics [(e.g., building motion salience on top of De Brecht and Saiki Neural Networks 19:1467-1474, (2006)]. The resulting network elicits strong transient responses to moving objects and reaches stability within a biologically plausible time interval. The responses are statistically different comparing between earlier and later motion neural activity; and between moving and non-moving objects. We demonstrate the network on a number of synthetic and real dynamical movie examples. We show that the model captures the motion saliency asymmetry phenomenon. In addition, the motion salience computation enables sudden-onset moving objects that are less salient in the static scene to rise above others. Finally, we include strong consideration for the neural latencies, the Lyapunov stability, and the neural properties being reproduced by the model.

Original languageEnglish (US)
Pages (from-to)141-160
Number of pages20
JournalBiological Cybernetics
Volume107
Issue number2
DOIs
StatePublished - Apr 2013
Externally publishedYes

Fingerprint

Dynamic analysis
Neural networks
Transient analysis
Neural Networks (Computer)
Motion Pictures

Keywords

  • Asymmetry phenomenon
  • Lyapunov stability
  • Motion saliency
  • Neural latency
  • Neural network
  • Saliency map
  • Synaptic depression
  • Visual attention

ASJC Scopus subject areas

  • Biotechnology
  • Computer Science(all)

Cite this

A saliency-based bottom-up visual attention model for dynamic scenes analysis. / Ramirez-Moreno, David F.; Schwartz, Odelia; Ramirez-Villegas, Juan F.

In: Biological Cybernetics, Vol. 107, No. 2, 04.2013, p. 141-160.

Research output: Contribution to journalArticle

Ramirez-Moreno, David F. ; Schwartz, Odelia ; Ramirez-Villegas, Juan F. / A saliency-based bottom-up visual attention model for dynamic scenes analysis. In: Biological Cybernetics. 2013 ; Vol. 107, No. 2. pp. 141-160.
@article{2e4bc5fa925a4ae49cd22b8d0646e3ff,
title = "A saliency-based bottom-up visual attention model for dynamic scenes analysis",
abstract = "This work proposes a model of visual bottom-up attention for dynamic scene analysis. Our work adds motion saliency calculations to a neural network model with realistic temporal dynamics [(e.g., building motion salience on top of De Brecht and Saiki Neural Networks 19:1467-1474, (2006)]. The resulting network elicits strong transient responses to moving objects and reaches stability within a biologically plausible time interval. The responses are statistically different comparing between earlier and later motion neural activity; and between moving and non-moving objects. We demonstrate the network on a number of synthetic and real dynamical movie examples. We show that the model captures the motion saliency asymmetry phenomenon. In addition, the motion salience computation enables sudden-onset moving objects that are less salient in the static scene to rise above others. Finally, we include strong consideration for the neural latencies, the Lyapunov stability, and the neural properties being reproduced by the model.",
keywords = "Asymmetry phenomenon, Lyapunov stability, Motion saliency, Neural latency, Neural network, Saliency map, Synaptic depression, Visual attention",
author = "Ramirez-Moreno, {David F.} and Odelia Schwartz and Ramirez-Villegas, {Juan F.}",
year = "2013",
month = "4",
doi = "10.1007/s00422-012-0542-2",
language = "English (US)",
volume = "107",
pages = "141--160",
journal = "Biological Cybernetics",
issn = "0340-1200",
publisher = "Springer Verlag",
number = "2",

}

TY - JOUR

T1 - A saliency-based bottom-up visual attention model for dynamic scenes analysis

AU - Ramirez-Moreno, David F.

AU - Schwartz, Odelia

AU - Ramirez-Villegas, Juan F.

PY - 2013/4

Y1 - 2013/4

N2 - This work proposes a model of visual bottom-up attention for dynamic scene analysis. Our work adds motion saliency calculations to a neural network model with realistic temporal dynamics [(e.g., building motion salience on top of De Brecht and Saiki Neural Networks 19:1467-1474, (2006)]. The resulting network elicits strong transient responses to moving objects and reaches stability within a biologically plausible time interval. The responses are statistically different comparing between earlier and later motion neural activity; and between moving and non-moving objects. We demonstrate the network on a number of synthetic and real dynamical movie examples. We show that the model captures the motion saliency asymmetry phenomenon. In addition, the motion salience computation enables sudden-onset moving objects that are less salient in the static scene to rise above others. Finally, we include strong consideration for the neural latencies, the Lyapunov stability, and the neural properties being reproduced by the model.

AB - This work proposes a model of visual bottom-up attention for dynamic scene analysis. Our work adds motion saliency calculations to a neural network model with realistic temporal dynamics [(e.g., building motion salience on top of De Brecht and Saiki Neural Networks 19:1467-1474, (2006)]. The resulting network elicits strong transient responses to moving objects and reaches stability within a biologically plausible time interval. The responses are statistically different comparing between earlier and later motion neural activity; and between moving and non-moving objects. We demonstrate the network on a number of synthetic and real dynamical movie examples. We show that the model captures the motion saliency asymmetry phenomenon. In addition, the motion salience computation enables sudden-onset moving objects that are less salient in the static scene to rise above others. Finally, we include strong consideration for the neural latencies, the Lyapunov stability, and the neural properties being reproduced by the model.

KW - Asymmetry phenomenon

KW - Lyapunov stability

KW - Motion saliency

KW - Neural latency

KW - Neural network

KW - Saliency map

KW - Synaptic depression

KW - Visual attention

UR - http://www.scopus.com/inward/record.url?scp=84876288269&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876288269&partnerID=8YFLogxK

U2 - 10.1007/s00422-012-0542-2

DO - 10.1007/s00422-012-0542-2

M3 - Article

C2 - 23314730

AN - SCOPUS:84876288269

VL - 107

SP - 141

EP - 160

JO - Biological Cybernetics

JF - Biological Cybernetics

SN - 0340-1200

IS - 2

ER -