Perceptual Organization as a Basis for Visual Recognition

Evidence is prescnled showiug that bottom-up grouping of im&ge features is usually prerequisite to the recognition and in terprc tation of images. WC describe three functions of Ihcse groupings: 1) scgmcnlation, 2) three-dimcnsiomrl intcrpreta, tion, and 3) stable descriptions for accessjug object models. Scvernl principlcc J are hypothesized for dctermining which image relations should be formed: relations are significant to the extent that they are unlikely to have arisen by accident from the surrounding distribution of features, relations can only be formed where there are few altcrnatives within tilt same proximity, and relations must be based on properties which are invariant over a range of imaging conditions. Using these principles we develop an algorithm for curve segmentation which detects significant structure at multiple rcsofutions, including the linking of segments on the basis of curvilinearity. The algorithm is a. ble to detect structures \vhich no single-resolution algorithm could detect. Its j>crformance is demonstrated OR synthetic and natural image data.

Authors

• David G. Lowe Thomas O. Binford

Keywords

No keywords are indexed for this paper.