Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Neural network model for extracting optic flow.

Kazuya Tohyama1, Kunihiko Fukushima

  • 1Tokyo University of Technology Hachioji, Tokyo 192-0982, Japan. tmkz@mf.teu.ac.jp

Neural Networks : the Official Journal of the International Neural Network Society
|August 23, 2005
PubMed
Summary

This study proposes a novel neural network model for optic flow analysis, inspired by vector-field calculus. The model effectively simulates how neurons in the medial superior temporal area (MST) process visual motion cues during navigation.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Efficient IntVec: High recognition rate with reduced computational cost.

Neural networks : the official journal of the International Neural Network Society·2019
Same author

Introduction.

International journal of neural systems·2018
Same author

Margined winner-take-all: New learning rule for pattern recognition.

Neural networks : the official journal of the International Neural Network Society·2017
Same author

Training multi-layered neural network neocognitron.

Neural networks : the official journal of the International Neural Network Society·2013
Same author

Artificial vision by multi-layered neural networks: neocognitron and its advances.

Neural networks : the official journal of the International Neural Network Society·2012
Same author

Increasing robustness against background noise: visual pattern recognition by a neocognitron.

Neural networks : the official journal of the International Neural Network Society·2011

Area of Science:

  • Neuroscience
  • Computational Neuroscience
  • Computer Vision

Background:

  • Optic flow, the apparent motion of objects in a visual scene, is crucial for navigation.
  • Neurons in the medial superior temporal area (MST) of macaque monkeys exhibit large receptive fields and analyze optic flows, responding selectively to various motion types like rotation, expansion/contraction, and planar motion.
  • Many MST cells demonstrate position-invariant responses, maintaining signal integrity despite shifts in the optic flow's center.

Purpose of the Study:

  • To propose a biologically plausible neural network model for optic flow extraction based on the vector-field hypothesis.
  • To investigate how hierarchical neural networks can replicate the complex optic flow processing observed in the brain.

Main Methods:

  • A hierarchical neural network model was developed, comprising layers simulating the retina, V1 (primary visual cortex), MT (middle temporal area), and MST.

Related Experiment Videos

  • V1 cells were modeled to measure local velocities.
  • MT cells were designed to extract both absolute and relative velocities through antagonistic inputs.
  • MST cells integrated signals from MT cells to achieve selective responses to different optic flow patterns.

    • Main Results:

    • The proposed neural network model successfully extracted optic flows.
    • Computer simulations demonstrated that the network could replicate various neurophysiological experimental findings.
    • The model showed that a hierarchical structure processing velocity information is sufficient for complex optic flow analysis.

    Conclusions:

    • The vector-field hypothesis provides a viable framework for developing computational models of optic flow processing.
    • The proposed hierarchical neural network effectively explains the functional properties of MST neurons in analyzing optic flow.
    • This model offers insights into the neural mechanisms underlying visual motion perception and navigation.