3D deep shape descriptor

Yi Fang, Jin Xie, Guoxian Dai, Meng Wang, Fan Zhu, Tiantian Xu, Edward Wong

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

Abstract

Shape descriptor is a concise yet informative representation that provides a 3D object with an identification as a member of some category. We have developed a concise deep shape descriptor to address challenging issues from ever-growing 3D datasets in areas as diverse as engineering, medicine, and biology. Specifically, in this paper, we developed novel techniques to extract concise but geometrically informative shape descriptor and new methods of defining Eigen-shape descriptor and Fisher-shape descriptor to guide the training of a deep neural network. Our deep shape descriptor tends to maximize the inter-class margin while minimize the intra-class variance. Our new shape descriptor addresses the challenges posed by the high complexity of 3D model and data representation, and the structural variations and noise present in 3D models. Experimental results on 3D shape retrieval demonstrate the superior performance of deep shape descriptor over other state-of-the-art techniques in handling noise, incompleteness and structural variations.

Original languageEnglish (US)
Title of host publicationIEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015
PublisherIEEE Computer Society
Pages2319-2328
Number of pages10
Volume07-12-June-2015
ISBN (Print)9781467369640
DOIs
StatePublished - Oct 14 2015
EventIEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015 - Boston, United States
Duration: Jun 7 2015Jun 12 2015

Other

OtherIEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015
CountryUnited States
CityBoston
Period6/7/156/12/15

Fingerprint

Medicine
Deep neural networks

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Fang, Y., Xie, J., Dai, G., Wang, M., Zhu, F., Xu, T., & Wong, E. (2015). 3D deep shape descriptor. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015 (Vol. 07-12-June-2015, pp. 2319-2328). [7298845] IEEE Computer Society. https://doi.org/10.1109/CVPR.2015.7298845

3D deep shape descriptor. / Fang, Yi; Xie, Jin; Dai, Guoxian; Wang, Meng; Zhu, Fan; Xu, Tiantian; Wong, Edward.

IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015. Vol. 07-12-June-2015 IEEE Computer Society, 2015. p. 2319-2328 7298845.

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

Fang, Y, Xie, J, Dai, G, Wang, M, Zhu, F, Xu, T & Wong, E 2015, 3D deep shape descriptor. in IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015. vol. 07-12-June-2015, 7298845, IEEE Computer Society, pp. 2319-2328, IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015, Boston, United States, 6/7/15. https://doi.org/10.1109/CVPR.2015.7298845
Fang Y, Xie J, Dai G, Wang M, Zhu F, Xu T et al. 3D deep shape descriptor. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015. Vol. 07-12-June-2015. IEEE Computer Society. 2015. p. 2319-2328. 7298845 https://doi.org/10.1109/CVPR.2015.7298845
Fang, Yi ; Xie, Jin ; Dai, Guoxian ; Wang, Meng ; Zhu, Fan ; Xu, Tiantian ; Wong, Edward. / 3D deep shape descriptor. IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015. Vol. 07-12-June-2015 IEEE Computer Society, 2015. pp. 2319-2328
@inproceedings{13398906d1ae4389b4534fc2bcfdeb0c,
title = "3D deep shape descriptor",
abstract = "Shape descriptor is a concise yet informative representation that provides a 3D object with an identification as a member of some category. We have developed a concise deep shape descriptor to address challenging issues from ever-growing 3D datasets in areas as diverse as engineering, medicine, and biology. Specifically, in this paper, we developed novel techniques to extract concise but geometrically informative shape descriptor and new methods of defining Eigen-shape descriptor and Fisher-shape descriptor to guide the training of a deep neural network. Our deep shape descriptor tends to maximize the inter-class margin while minimize the intra-class variance. Our new shape descriptor addresses the challenges posed by the high complexity of 3D model and data representation, and the structural variations and noise present in 3D models. Experimental results on 3D shape retrieval demonstrate the superior performance of deep shape descriptor over other state-of-the-art techniques in handling noise, incompleteness and structural variations.",
author = "Yi Fang and Jin Xie and Guoxian Dai and Meng Wang and Fan Zhu and Tiantian Xu and Edward Wong",
year = "2015",
month = "10",
day = "14",
doi = "10.1109/CVPR.2015.7298845",
language = "English (US)",
isbn = "9781467369640",
volume = "07-12-June-2015",
pages = "2319--2328",
booktitle = "IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - 3D deep shape descriptor

AU - Fang, Yi

AU - Xie, Jin

AU - Dai, Guoxian

AU - Wang, Meng

AU - Zhu, Fan

AU - Xu, Tiantian

AU - Wong, Edward

PY - 2015/10/14

Y1 - 2015/10/14

N2 - Shape descriptor is a concise yet informative representation that provides a 3D object with an identification as a member of some category. We have developed a concise deep shape descriptor to address challenging issues from ever-growing 3D datasets in areas as diverse as engineering, medicine, and biology. Specifically, in this paper, we developed novel techniques to extract concise but geometrically informative shape descriptor and new methods of defining Eigen-shape descriptor and Fisher-shape descriptor to guide the training of a deep neural network. Our deep shape descriptor tends to maximize the inter-class margin while minimize the intra-class variance. Our new shape descriptor addresses the challenges posed by the high complexity of 3D model and data representation, and the structural variations and noise present in 3D models. Experimental results on 3D shape retrieval demonstrate the superior performance of deep shape descriptor over other state-of-the-art techniques in handling noise, incompleteness and structural variations.

AB - Shape descriptor is a concise yet informative representation that provides a 3D object with an identification as a member of some category. We have developed a concise deep shape descriptor to address challenging issues from ever-growing 3D datasets in areas as diverse as engineering, medicine, and biology. Specifically, in this paper, we developed novel techniques to extract concise but geometrically informative shape descriptor and new methods of defining Eigen-shape descriptor and Fisher-shape descriptor to guide the training of a deep neural network. Our deep shape descriptor tends to maximize the inter-class margin while minimize the intra-class variance. Our new shape descriptor addresses the challenges posed by the high complexity of 3D model and data representation, and the structural variations and noise present in 3D models. Experimental results on 3D shape retrieval demonstrate the superior performance of deep shape descriptor over other state-of-the-art techniques in handling noise, incompleteness and structural variations.

UR - http://www.scopus.com/inward/record.url?scp=84959185649&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959185649&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2015.7298845

DO - 10.1109/CVPR.2015.7298845

M3 - Conference contribution

AN - SCOPUS:84959185649

SN - 9781467369640

VL - 07-12-June-2015

SP - 2319

EP - 2328

BT - IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015

PB - IEEE Computer Society

ER -