Sage Journals: Discover world-class research

Abstract

Two experiments tested the predictions of a recent theory of object recognition (Recognition-By-Components or RBC). The first experiment supports RBC's predictions that recognition is accomplished by ascertaining the object's 3D components in their specified relations by utilizing certain nonaccidental 2D image properties. When object components were rendered nonrecoverable by deleting their nonaccidental properties recognition in an object naming task was interfered with more than when an equal amount of contour was deleted without erasing such properties. The second experiment supported RBC's assumption that vertices constitute one of these important nonaccidental properties. In a naming task objects were recognized more accurately when contour was deleted from midsegment than when an equal amount was deleted from the objects vertices.

Get full access to this article

View all access options for this article.

References

Biederman

(1985). Human image understanding: Recent research and a theory. Computer Vision, Graphics, and Image Processing, 32, 29–73.

Gollin

E. S.

(1961). Further studies of visual recognition of incomplete objects. Perceptual and Motor Skills, 13 307–314.

Lowe

(1984). Perceptual organization and visuzl recognition. Unpublish doctoral dissertation, Department of Computer Science, Stanford University.

Marr

Nishihara

H. K.

(1978). Repressentation and recognition of the spatial organization of three-dimensional shapes. Proc. R. Soc. Lond. B 200, 269–294.

Pentland

A. P.

(1985). Perceptual organization and the representation of natural form. SRI Artificial Intelligence Center Technical Note 357.

Spitz

H. H.

Borland

M. D.

(1971). Redunancy in line drawings of familiar objects: Effects of age and intelligence. Cognitive Psychology, 2, 196–205.

Tests of a Theory of Human Image Understanding: Part II. The Perception of Degraded Objects

Abstract

Get full access to this article

References