We describe a system for obtaining a ''generic'' parts-based 3D object
representation. We use range image data as the input, obtaining a 3D
object representation based on 12 geon-like 3D part primitives as the
output. The 3D parts-based representation consists of parts detected i
n the image and their identities. Unlike previous work, we do not make
simplifying assumptions such as the availability of perfect line draw
ings, perfect segmentation, or manual segmentation. We propose a novel
method of specifying ''generic'' 3D parts, i.e., by means of surface
adjacency graphs (SAGs). Using the SAGs, we derive an extremely compac
t multi-view representation of the part primitives, consisting of a to
tal of only 74 views for all 12 primitives. Based on the multi-view re
presentation of parts, we present a method of performing part segmenta
tion from range images, given a good surface segmentation. This method
for part segmentation is more general than common approaches based on
Hoffman and Richards' ''principle of transversality.'' We present two
approaches for identifying the parts as one of the 12 3D part primiti
ves. The first approach applies statistical pattern classification met
hods using parameters estimated by superquadric fitting. Five features
derived from the estimated superquadric parameters are used to distin
guish between the 12 part primitives. Classification error rates are e
stimated for k-nearest-neighbor and binary tree classifiers, for real
as well as for synthetic range images. The second approach for part id
entification draws inferences from the distribution of angles between
surface normals and the principal axis of a part. We show that intensi
ty data can be used to recover from some misclassifications yielded by
the purely range-based methods of part identification. A simple test
is applied to check the concavity or convexity of the part silhouette
in the intensity image. This serves as a reliable test of whether the
part axis is straight or curved. Results of part segmentation and iden
tification are presented for real range images of several multi-part o
bjects. Our system successfully performs part segmentation and identif
ies the parts. (C) 1994 Academic Press, Inc.