Visual motion compensation using a neural shifter circuit

Author

Olshausen, Bruno A.

Author_Institution

Comput. & Neural Syst. Program, California Inst. of Technol., Pasadena, CA, USA

fYear

1990

fDate

4-7 Nov 1990

Firstpage

829

Lastpage

831

Abstract

It was recently proposed that a neural shifter circuit could be responsible for compensating for image motion so as to produce an unblurred image in visual cortex. An analysis of the motion blur problem in the frequency domain is presented, and it is shown how a neural shifter circuit behaves as a filter within this domain in order to extract spatial information from moving images. From this analysis it is possible to formally compare the shifter circuit to other computational strategies for reducing motion blur, such as spatio-temporal filters. A control systems analysis of the shifter circuit that allows one to predict how neural time delays affect the performance of the system is provided. The results of a computer simulation of a shifter circuit are presented

Keywords

frequency-domain analysis; neural nets; physiological models; spatial filters; vision; frequency domain; image motion compensation; motion blur; neural shifter circuit; neural time delays; spatio-temporal filters; vision; visual cortex; Circuits; Data mining; Frequency domain analysis; Image analysis; Image motion analysis; Information analysis; Information filtering; Information filters; Motion analysis; Motion compensation;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Man and Cybernetics, 1990. Conference Proceedings., IEEE International Conference on

Conference_Location

Los Angeles, CA

Print_ISBN

0-87942-597-0

Type

conf

DOI

10.1109/ICSMC.1990.142238

Filename

142238