creators_name: Hillenbrand, Ulrich
creators_name: Hirzinger, Gerd
type: confpaper
datestamp: 2002-08-09
lastmod: 2011-03-11 08:54:58
metadata_visibility: show
title: Probabilistic Search for Object Segmentation and Recognition
ispublished: pub
subjects: comp-sci-mach-vis
subjects: comp-sci-mach-learn
subjects: comp-sci-robot
full_text_status: public
keywords: scene analysis, object recognition, grouping, segmentation, surface geometry, statistical modeling, range-data processing, searching, optimization
abstract: The problem of searching for a model-based scene interpretation is analyzed within a probabilistic framework. Object models are formulated as generative models for range data of the scene. A new statistical criterion, the truncated object probability, is introduced to infer an optimal sequence of object hypotheses to be evaluated for their match to the data. The truncated probability is partly determined by prior knowledge of the objects and partly learned from data. Some experiments on sequence quality and object segmentation and recognition from stereo data are presented. The article recovers classic concepts from object recognition (grouping, geometric hashing, alignment) from the probabilistic perspective and adds insight into the optimal ordering of object hypotheses for evaluation. Moreover, it introduces point-relation densities, a key component of the truncated probability, as statistical models of local surface shape.
date: 2002
date_type: published
pagerange: 791-806
refereed: TRUE
referencetext: \bibitem{Ballard81}
{\sc Ballard, D.~H.}
\newblock Generalizing the {Hough} transform to detect arbitrary shapes.
\newblock {\em Pattern Recognit. 13\/} (1981), 111--122.

\bibitem{Besl&Jain85a}
{\sc Besl, P., and Jain, R.}
\newblock Intrinsic and extrinsic surface characteristics.
\newblock In {\em Proc.\ IEEE Conf.\ Comp.\ Vision Patt.\ Recogn.\/} (1985),
  pp.~226--233.

\bibitem{Besl&Jain85b}
{\sc Besl, P., and Jain, R.}
\newblock Three-dimensional object recognition.
\newblock {\em ACM Comput.\ Surveys 17\/} (1985), 75--145.

\bibitem{Faugeras&Hebert87}
{\sc Faugeras, O.~D., and Hebert, M.}
\newblock The representation, recognition, and positioning of {3-D} shapes from
  range data.
\newblock In {\em Three-Dimensional Machine Vision}, T.~Kanade, Ed. Kluwer
  Academic Publishers, 1987, pp.~301--353.

\bibitem{Fischler&Bolls81}
{\sc Fischler, M., and Bolles, R.}
\newblock Random sample consensus: {A} paradigm for model fitting with
  application to image analysis and automated cartography.
\newblock {\em Comm.\ Assoc.\ Comp.\ Mach. 24\/} (1981), 381--395.

\bibitem{Fleuret&Geman99}
{\sc Fleuret, F., and Geman, D.}
\newblock Graded learning for object detection.
\newblock In {\em Proc.\ IEEE Workshop Stat.\ Comput.\ Theo.\ Vision\/} (1999).

\bibitem{Goad83}
{\sc Goad, C.}
\newblock Special purpose automatic programming for {3D} model-based vision.
\newblock In {\em Proc.\ ARPA Image Understanding Workshop\/} (1983).

\bibitem{Grimson90}
{\sc Grimson, W. E.~L.}
\newblock {\em Object Recognition by Computer: {The} Role of Geometric
  Constraints}.
\newblock MIT Press, Cambridge, MA, 1990.

\bibitem{Huttenlocher&Ullman87}
{\sc Huttenlocher, D.~P., and Ullman, S.}
\newblock Object recognition using alignment.
\newblock In {\em Proc.\ Intern.\ Conf.\ Comput.\ Vision\/} (1987),
  pp.~102--111.

\bibitem{Huttenlocher&Ullman90}
{\sc Huttenlocher, D.~P., and Ullman, S.}
\newblock Recognizing solid objects by alignment with an image.
\newblock {\em Int.\ J.\ Computer Vision 5\/} (1990), 195--212.

\bibitem{Clemens&Jacobs91}
{\sc Jacobs, D., and Clemens, D.}
\newblock Space and time bounds on indexing {3-D} models from {2-D} images.
\newblock {\em IEEE Trans.\ Patt.\ Anal.\ Mach.\ Intell. 13\/} (1991),
  1007--1017.

\bibitem{Kuno_etal91}
{\sc Kuno, Y., Okamoto, Y., and Okada, S.}
\newblock Robot vision using a feature search strategy generated from a {3-D}
  object model.
\newblock {\em IEEE Trans.\ Patt.\ Anal.\ Machine Intell. 13\/} (1991),
  1085--1097.

\bibitem{Lamdan&Wolfson88}
{\sc Lamdan, Y., and Wolfson, H.~J.}
\newblock Geometric hashing: a general and efficient model-based recognition
  scheme.
\newblock In {\em Proc.\ Intern.\ Conf.\ Comput.\ Vision\/} (1988),
  pp.~238--249.

\bibitem{Lowe85}
{\sc Lowe, D.~G.}
\newblock {\em Perceptual Organization and Visual Recognition}.
\newblock Kluwer, Boston, 1985.

\bibitem{Lowe90}
{\sc Lowe, D.~G.}
\newblock Visual recognition as probabilistic inference from spatial relations.
\newblock In {\em {AI} and the Eye}, A.~Blake and T.~Troscianko, Eds. John
  Wiley and Sons Ltd., New York, 1990, pp.~261--279.

\bibitem{Moss_etal99}
{\sc Moss, S., Wilson, R.~C., and Hancock, E.~R.}
\newblock A mixture model for pose clustering.
\newblock {\em Patt.\ Recogn.\ Let. 20\/} (1999), 1093--1101.

\bibitem{Stockman87}
{\sc Stockmann, G.}
\newblock Object recognition and localization via pose clustering.
\newblock {\em CVGIP 40\/} (1987), 361--387.

\bibitem{Stokely&YouWu92}
{\sc Stokely, E.~M., and Wu, S.~Y.}
\newblock Surface parameterization and curvature measurement of arbitrary {3-D}
  objects: {Five} practical methods.
\newblock {\em IEEE Trans.\ Patt.\ Anal.\ Mach.\ Intell. 14\/} (1992),
  833--840.

\bibitem{Torr&Zisserman00}
{\sc Torr, P. H.~S., and Zisserman, A.}
\newblock {MLESAC}: {A} new robust estimator with application to estimating
  image geometry.
\newblock {\em Comput.\ Vision Image Understanding 78\/} (2000), 138--156.

\bibitem{Wolfson90}
{\sc Wolfson, H.~J.}
\newblock Model-based object recognition by geometric hashing.
\newblock In {\em Proc.\ Eur.\ Conf.\ Comput.\ Vision\/} (1990), pp.~526--536.
citation:   Hillenbrand, Dr. Ulrich and Hirzinger, Prof. Dr. Gerd  (2002) Probabilistic Search for Object Segmentation and Recognition.  [Conference Paper]     
document_url: http://cogprints.org/2393/1/Hillenbrand_Hirzinger_02.pdf