Cepstral Filtering on a Columnar Image Architecture: A Fast Algorithm for Binocular Stereo Segmentation

Yehezkel Yeshurun, Eric L. Schwartz

Research output: Contribution to journalArticlepeer-review

80 Scopus citations

Abstract

Many primate visual cortex architectures (including the human) have a prominent feature responsible for the mixing of left and right eye visual data: ocular dominance columns represent thin (about 5-10 minutes of arc) strips of alternating left and right eye input to the brain. In the present paper we show that such an architecture, when operated upon with a cepstral filter, provides a strong cue for binocular stereopsis. Specifically, the vector of binocular disparity may be easily identified in the output of the (columnar based) cepstral filter. This algorithm is illustrated with application to a random dot stereogram and to natural images. We suggest that this provides a fast algorithm for stereo segmentation, in a machine vision context. In a biological context, this may provide a computational rationale for the existence of columnar systems, both with regard to ocular mixing, and to other visual modalities which have a columnar architecture.

Original languageEnglish
Pages (from-to)759-767
Number of pages9
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume11
Issue number7
DOIs
StatePublished - Jul 1989
Externally publishedYes

Funding

FundersFunder number
Air Force Office of Scientific Research85-0341

    Fingerprint

    Dive into the research topics of 'Cepstral Filtering on a Columnar Image Architecture: A Fast Algorithm for Binocular Stereo Segmentation'. Together they form a unique fingerprint.

    Cite this