0
Journal of Mechanical Design | Volume 139 | Issue 1 | Technical Brief
Technical Brief

Probabilistic Factor Graph Based Approach for Automatic Material Assignments to Three-Dimensional Objects

[+] Author and Article Information
Binbin Zhang

Manufacturing and Design Lab (MADLab),
Department of Mechanical and Aerospace Engineering,
University at Buffalo,
Buffalo, NY 14260
e-mail: bzhang25@buffalo.edu

Rahul Rai

Manufacturing and Design Lab (MADLab),
Department of Mechanical and Aerospace Engineering,
University at Buffalo,
Buffalo, NY 14260
e-mail: rahulrai@buffalo.edu

1Corresponding author.

Contributed by the Design Theory and Methodology Committee of ASME for publication in the JOURNAL OF MECHANICAL DESIGN. Manuscript received September 9, 2015; final manuscript received September 19, 2016; published online October 14, 2016. Assoc. Editor: Kristina Shea.

J. Mech. Des 139(1), 014501 (Oct 14, 2016) (8 pages) Paper No: MD-15-1635; doi: 10.1115/1.4034838 History: Received September 09, 2015; Revised September 19, 2016
Article
References
Figures

There are strong correlations between material assignment, shape, and functionality of a part in an overall product/assembly. However, these strong correlations are rarely exploited for automated material assignment. We present a probabilistic graphical model to assign materials to the parts (components) of a 3D object (assembly) by identifying the relations between shape, functionality, and material of the parts. By learning the context-dependent correlation with supervision from a set of objects and their segmented parts, the learned model can be used to assign engineering materials to the parts of a query object. Our primary contributions are (a) the engineering materials definition and assignment and (b) assigning engineering materials based on the behavior and form of the parts in the object. The performance of the proposed computational approach is demonstrated by the results of material assignment on various query objects with prespecified engineering performance requirements.
FIGURES IN THIS ARTICLE
<>
Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.

References

1
Jain, A. , Thormählen, T. , Ritschel, T. , and Seidel, H.-P. , 2012, “ Material Memex: Automatic Material Suggestions for 3D Objects,” ACM Trans. Graphics, 31(6), pp. 143–150. [CrossRef]
2
Ullman, D. G. , 1992, The Mechanical Design Process, Vol. 2, McGraw-Hill, New York.
3
Malisiewicz, T. , and Efros, A. , 2009, “ Beyond Categories: The Visual Memex Model for Reasoning About Object Relationships,” Advances in Neural Information Processing Systems, pp. 1222–1230.
4
Kollar, D. , and Friedman, N. , 2009, Probabilistic Graphical Models: Principles and Techniques, MIT Press, Cambridge, MA.
5
Shilane, P. , Min, P. , Kazhdan, M. , and Funkhouser, T. , 2004, “ The Princeton Shape Benchmark,” Shape Modeling Applications, IEEE, Los Alamitos, CA, June 7–9, pp. 167–178.
6
Gal, R. , and Cohen-Or, D. , 2006, “ Salient Geometric Features for Partial Shape Matching and Similarity,” ACM Trans. Graphics, 25(1), pp. 130–150. [CrossRef]
7
Chen, D.-Y. , Tian, X.-P. , Shen, Y.-T. , and Ouhyoung, M. , 2003, “ On Visual Similarity Based 3D Model Retrieval,” Computer Graphics Forum, Vol. 22, Wiley Online Library, Malden, MA, pp. 223–232.
8
Ashby, M. F. , and Cebon, D. , 1993, “ Materials Selection in Mechanical Design,” J. Phys. IV, 3(C7), pp. C7–C1.
9
Parametric Technology Corporation, 2016, “ Creo,” PTC Software, Needham, MA, accessed June 3, 2016, http://www.ptc.com/cad/creo
10
Mooij, J. M. , and Kappen, H. J. , 2007, “ Sufficient Conditions for Convergence of the Sum–Product Algorithm,” IEEE Trans. Inf. Theory, 53(12), pp. 4422–4437. [CrossRef]
11
Kschischang, F. R. , Frey, B. J. , and Loeliger, H.-A. , 2001, “ Factor Graphs and the Sum-Product Algorithm,” IEEE Trans. Inf. Theory, 47(2), pp. 498–519. [CrossRef]

Figures

Grahic Jump Location
Fig. 1

Automatic material assignment: office chair. (a) Input 3D model with desired behavior and (b) result of our automatic material assignment process based on shape and behavior.

+Automatic material assignment: office chair. (a) Input 3D model with desired behavior and (b) result of our automatic material assignment process based on shape and behavior.
View Large  |  Save Figure  |  Download Slide (.ppt)
Automatic material assignment: office chair. (a) Input 3D model with desired behavior and (b) result of our automatic material assignment process based on shape and behavior.
Grahic Jump Location
Fig. 2

Technical approach overview

+Technical approach overview
View Large  |  Save Figure  |  Download Slide (.ppt)
Technical approach overview
Grahic Jump Location
Fig. 3

Segmentation and database memex graph examples

+Segmentation and database memex graph examples
View Large  |  Save Figure  |  Download Slide (.ppt)
Segmentation and database memex graph examples
Grahic Jump Location
Fig. 4

Steps for defining the behavior of a database object

+Steps for defining the behavior of a database object
View Large  |  Save Figure  |  Download Slide (.ppt)
Steps for defining the behavior of a database object
Grahic Jump Location
Fig. 5

Database object: office chair. (a) Material allocation and the maximum allowable loads on the seat and seat back and (b) Von Mises stress (the yield strength of alloy steel is 6.20422 × 108 N/m2).

+Database object: office chair. (a) Material allocation and the maximum allowable loads on the seat and seat back and (b) Von Mises stress (the yield strength of alloy steel is 6.20422 × 108 N/m2).
View Large  |  Save Figure  |  Download Slide (.ppt)
Database object: office chair. (a) Material allocation and the maximum allowable loads on the seat and seat back and (b) Von Mises stress (the yield strength of alloy steel is 6.20422 × 108 N/m2).
Grahic Jump Location
Fig. 6

Steps for defining the behavior of the query object

+Steps for defining the behavior of the query object
View Large  |  Save Figure  |  Download Slide (.ppt)
Steps for defining the behavior of the query object
Grahic Jump Location
Fig. 7

Verification process for the query object

+Verification process for the query object
View Large  |  Save Figure  |  Download Slide (.ppt)
Verification process for the query object
Grahic Jump Location
Fig. 8

Query memex graph and factor graph example. (This is an extension of Jain et al. [1]. For the query memex graph, the behavioral similarity edge is added, which takes the behavior aspect of parts into consideration. Correspondingly, in the factor graph, a pairwise factor ψB is included.)

+Query memex graph and factor graph example. (This is an extension of Jain et al. [1]. For the query memex graph, the behavioral similarity edge is added, which takes the behavior aspect of parts into consideration. Correspondingly, in the factor graph, a pairwise factor ψB is included.)
View Large  |  Save Figure  |  Download Slide (.ppt)
Query memex graph and factor graph example. (This is an extension of Jain et al. [1]. For the query memex graph, the behavioral similarity edge is added, which takes the behavior aspect of parts into consideration. Correspondingly, in the factor graph, a pairwise factor ψB is included.)
Grahic Jump Location
Fig. 9

The marginal probability values for assigning different materials to chair seat, one-layer house roof, and two-layer house roof

+The marginal probability values for assigning different materials to chair seat, one-layer house roof, and two-layer house roof
View Large  |  Save Figure  |  Download Slide (.ppt)
The marginal probability values for assigning different materials to chair seat, one-layer house roof, and two-layer house roof
Grahic Jump Location
Fig. 10

Eight different query objects

+Eight different query objects
View Large  |  Save Figure  |  Download Slide (.ppt)
Eight different query objects
Grahic Jump Location
Fig. 11

Material assignment result for objects depicted in Fig. 10

+Material assignment result for objects depicted in Fig. 10
View Large  |  Save Figure  |  Download Slide (.ppt)
Material assignment result for objects depicted in Fig. 10
Grahic Jump Location
Fig. 12

Material assignment result on a bicycle model: (a) model without material and (b) result of automatic material assignment

+Material assignment result on a bicycle model: (a) model without material and (b) result of automatic material assignment
View Large  |  Save Figure  |  Download Slide (.ppt)
Material assignment result on a bicycle model: (a) model without material and (b) result of automatic material assignment

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Tooth Forms
Design and Application of the Worm Gear> Chapter 3
Visual Exploration of Numeric Data Using 3D Self Organizing Feature Maps
Intelligent Engineering Systems Through Artificial Neural Networks, Volume 17
Discovery of Useful Concepts Using the Hierarchy of Attributes and Concepts
Intelligent Engineering Systems Through Artificial Neural Networks, Volume 17
Generating Meaningful Rules Using Attribute Concept Hierarchy
Intelligent Engineering Systems through Artificial Neural Networks, Volume 16
Topographic Processing of Very Large Text Datasets
Intelligent Engineering Systems through Artificial Neural Networks Volume 18
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In