A 3D Shape Descriptor Based on Depth Complexity and Thickness Histograms

Wagner Schmitt, Jose L. Sotomayor, Alexandru Telea, Claudio T. Silva, Joao L D Comba

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Geometric models play a vital role in several fields, from the entertainment industry to scientific applications. To reduce the high cost of model creation, reusing existing models is the solution of choice. Model reuse is supported by content-based shape retrieval (CBR) techniques that help finding the desired models in massive repositories, many publicly available on the Internet. Key to efficient and effective CBR techniques are shape descriptors that accurately capture the characteristics of a shape and can discriminate between different shapes. We present a descriptor based on the distribution of two global features measured in a 3D shape, depth complexity and thickness, which respectively capture aspects of the geometry and topology of 3D shapes. The final descriptor, called DCTH (depth complexity and thickness histogram), is a 2D histogram that is invariant to the translation, rotation and scale of geometric shapes. We efficiently implement the DCTH on the GPU, allowing its use in real-time queries of large model databases. We validate the DCTH with the Princeton and Toyohashi Shape Benchmarks, containing 1815 and 10000 models respectively. Results show that DCTH can discriminate meaningful classes of these benchmarks and is fast to compute and robust against shape transformations and different levels of subdivision and smoothness.

Original languageEnglish (US)
Title of host publicationProceedings - 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015
PublisherIEEE Computer Society
Pages226-233
Number of pages8
Volume2015-October
ISBN (Print)9781467379625
DOIs
StatePublished - Oct 30 2015
Event28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015 - Salvador, Bahia, Brazil
Duration: Aug 26 2015Aug 29 2015

Other

Other28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015
CountryBrazil
CitySalvador, Bahia
Period8/26/158/29/15

Fingerprint

Topology
Internet
Geometry
Costs
Industry
Graphics processing unit

Keywords

  • Content-based retrieval
  • Depth complexity
  • Histograms
  • Shape analysis
  • Shape matching
  • Thickness

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Signal Processing
  • Software

Cite this

Schmitt, W., Sotomayor, J. L., Telea, A., Silva, C. T., & Comba, J. L. D. (2015). A 3D Shape Descriptor Based on Depth Complexity and Thickness Histograms. In Proceedings - 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015 (Vol. 2015-October, pp. 226-233). [7314568] IEEE Computer Society. https://doi.org/10.1109/SIBGRAPI.2015.51

A 3D Shape Descriptor Based on Depth Complexity and Thickness Histograms. / Schmitt, Wagner; Sotomayor, Jose L.; Telea, Alexandru; Silva, Claudio T.; Comba, Joao L D.

Proceedings - 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015. Vol. 2015-October IEEE Computer Society, 2015. p. 226-233 7314568.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Schmitt, W, Sotomayor, JL, Telea, A, Silva, CT & Comba, JLD 2015, A 3D Shape Descriptor Based on Depth Complexity and Thickness Histograms. in Proceedings - 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015. vol. 2015-October, 7314568, IEEE Computer Society, pp. 226-233, 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015, Salvador, Bahia, Brazil, 8/26/15. https://doi.org/10.1109/SIBGRAPI.2015.51
Schmitt W, Sotomayor JL, Telea A, Silva CT, Comba JLD. A 3D Shape Descriptor Based on Depth Complexity and Thickness Histograms. In Proceedings - 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015. Vol. 2015-October. IEEE Computer Society. 2015. p. 226-233. 7314568 https://doi.org/10.1109/SIBGRAPI.2015.51
Schmitt, Wagner ; Sotomayor, Jose L. ; Telea, Alexandru ; Silva, Claudio T. ; Comba, Joao L D. / A 3D Shape Descriptor Based on Depth Complexity and Thickness Histograms. Proceedings - 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015. Vol. 2015-October IEEE Computer Society, 2015. pp. 226-233
@inproceedings{7636a3eb76e041859af04a6abdacc637,
title = "A 3D Shape Descriptor Based on Depth Complexity and Thickness Histograms",
abstract = "Geometric models play a vital role in several fields, from the entertainment industry to scientific applications. To reduce the high cost of model creation, reusing existing models is the solution of choice. Model reuse is supported by content-based shape retrieval (CBR) techniques that help finding the desired models in massive repositories, many publicly available on the Internet. Key to efficient and effective CBR techniques are shape descriptors that accurately capture the characteristics of a shape and can discriminate between different shapes. We present a descriptor based on the distribution of two global features measured in a 3D shape, depth complexity and thickness, which respectively capture aspects of the geometry and topology of 3D shapes. The final descriptor, called DCTH (depth complexity and thickness histogram), is a 2D histogram that is invariant to the translation, rotation and scale of geometric shapes. We efficiently implement the DCTH on the GPU, allowing its use in real-time queries of large model databases. We validate the DCTH with the Princeton and Toyohashi Shape Benchmarks, containing 1815 and 10000 models respectively. Results show that DCTH can discriminate meaningful classes of these benchmarks and is fast to compute and robust against shape transformations and different levels of subdivision and smoothness.",
keywords = "Content-based retrieval, Depth complexity, Histograms, Shape analysis, Shape matching, Thickness",
author = "Wagner Schmitt and Sotomayor, {Jose L.} and Alexandru Telea and Silva, {Claudio T.} and Comba, {Joao L D}",
year = "2015",
month = "10",
day = "30",
doi = "10.1109/SIBGRAPI.2015.51",
language = "English (US)",
isbn = "9781467379625",
volume = "2015-October",
pages = "226--233",
booktitle = "Proceedings - 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - A 3D Shape Descriptor Based on Depth Complexity and Thickness Histograms

AU - Schmitt, Wagner

AU - Sotomayor, Jose L.

AU - Telea, Alexandru

AU - Silva, Claudio T.

AU - Comba, Joao L D

PY - 2015/10/30

Y1 - 2015/10/30

N2 - Geometric models play a vital role in several fields, from the entertainment industry to scientific applications. To reduce the high cost of model creation, reusing existing models is the solution of choice. Model reuse is supported by content-based shape retrieval (CBR) techniques that help finding the desired models in massive repositories, many publicly available on the Internet. Key to efficient and effective CBR techniques are shape descriptors that accurately capture the characteristics of a shape and can discriminate between different shapes. We present a descriptor based on the distribution of two global features measured in a 3D shape, depth complexity and thickness, which respectively capture aspects of the geometry and topology of 3D shapes. The final descriptor, called DCTH (depth complexity and thickness histogram), is a 2D histogram that is invariant to the translation, rotation and scale of geometric shapes. We efficiently implement the DCTH on the GPU, allowing its use in real-time queries of large model databases. We validate the DCTH with the Princeton and Toyohashi Shape Benchmarks, containing 1815 and 10000 models respectively. Results show that DCTH can discriminate meaningful classes of these benchmarks and is fast to compute and robust against shape transformations and different levels of subdivision and smoothness.

AB - Geometric models play a vital role in several fields, from the entertainment industry to scientific applications. To reduce the high cost of model creation, reusing existing models is the solution of choice. Model reuse is supported by content-based shape retrieval (CBR) techniques that help finding the desired models in massive repositories, many publicly available on the Internet. Key to efficient and effective CBR techniques are shape descriptors that accurately capture the characteristics of a shape and can discriminate between different shapes. We present a descriptor based on the distribution of two global features measured in a 3D shape, depth complexity and thickness, which respectively capture aspects of the geometry and topology of 3D shapes. The final descriptor, called DCTH (depth complexity and thickness histogram), is a 2D histogram that is invariant to the translation, rotation and scale of geometric shapes. We efficiently implement the DCTH on the GPU, allowing its use in real-time queries of large model databases. We validate the DCTH with the Princeton and Toyohashi Shape Benchmarks, containing 1815 and 10000 models respectively. Results show that DCTH can discriminate meaningful classes of these benchmarks and is fast to compute and robust against shape transformations and different levels of subdivision and smoothness.

KW - Content-based retrieval

KW - Depth complexity

KW - Histograms

KW - Shape analysis

KW - Shape matching

KW - Thickness

UR - http://www.scopus.com/inward/record.url?scp=84959367551&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959367551&partnerID=8YFLogxK

U2 - 10.1109/SIBGRAPI.2015.51

DO - 10.1109/SIBGRAPI.2015.51

M3 - Conference contribution

SN - 9781467379625

VL - 2015-October

SP - 226

EP - 233

BT - Proceedings - 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2015

PB - IEEE Computer Society

ER -