3D Capsule Networks for Object Classification from 3D Model Data

Ayesha Ahmad, Burak Kakillioglu, Senem Velipasalar

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

Abstract

Many of the existing object classification methods today rely on convolutional neural networks (CNNs), which are very successful in extracting features from the data. However, CNNs cannot sufficiently address the spatial relationship between features and require large amounts of data for training. In this paper, a new architecture is proposed for 3D object classification, which is an extension of the Capsule Networks (CapsNets) to 3D data. Our proposed 3D CapsNet architecture preserves the orientation and spatial relationship of the extracted features, and thus requires less data to train the network. We compare our approach with a ShapeNet inspired model, and show that our method provides performance improvement especially when training data size gets smaller. We also compare and evaluate several different versions of the 3D Capsnet architecture.

Original languageEnglish (US)
Title of host publicationConference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018
EditorsMichael B. Matthews
PublisherIEEE Computer Society
Pages2225-2229
Number of pages5
ISBN (Electronic)9781538692189
DOIs
StatePublished - Feb 19 2019
Event52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018 - Pacific Grove, United States
Duration: Oct 28 2018Oct 31 2018

Publication series

NameConference Record - Asilomar Conference on Signals, Systems and Computers
Volume2018-October
ISSN (Print)1058-6393

Conference

Conference52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018
CountryUnited States
CityPacific Grove
Period10/28/1810/31/18

Fingerprint

Neural networks

Keywords

  • 3D
  • Capsule networks
  • object classification

ASJC Scopus subject areas

  • Signal Processing
  • Computer Networks and Communications

Cite this

Ahmad, A., Kakillioglu, B., & Velipasalar, S. (2019). 3D Capsule Networks for Object Classification from 3D Model Data. In M. B. Matthews (Ed.), Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018 (pp. 2225-2229). [8645256] (Conference Record - Asilomar Conference on Signals, Systems and Computers; Vol. 2018-October). IEEE Computer Society. https://doi.org/10.1109/ACSSC.2018.8645256

3D Capsule Networks for Object Classification from 3D Model Data. / Ahmad, Ayesha; Kakillioglu, Burak; Velipasalar, Senem.

Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018. ed. / Michael B. Matthews. IEEE Computer Society, 2019. p. 2225-2229 8645256 (Conference Record - Asilomar Conference on Signals, Systems and Computers; Vol. 2018-October).

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

Ahmad, A, Kakillioglu, B & Velipasalar, S 2019, 3D Capsule Networks for Object Classification from 3D Model Data. in MB Matthews (ed.), Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018., 8645256, Conference Record - Asilomar Conference on Signals, Systems and Computers, vol. 2018-October, IEEE Computer Society, pp. 2225-2229, 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018, Pacific Grove, United States, 10/28/18. https://doi.org/10.1109/ACSSC.2018.8645256
Ahmad A, Kakillioglu B, Velipasalar S. 3D Capsule Networks for Object Classification from 3D Model Data. In Matthews MB, editor, Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018. IEEE Computer Society. 2019. p. 2225-2229. 8645256. (Conference Record - Asilomar Conference on Signals, Systems and Computers). https://doi.org/10.1109/ACSSC.2018.8645256
Ahmad, Ayesha ; Kakillioglu, Burak ; Velipasalar, Senem. / 3D Capsule Networks for Object Classification from 3D Model Data. Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018. editor / Michael B. Matthews. IEEE Computer Society, 2019. pp. 2225-2229 (Conference Record - Asilomar Conference on Signals, Systems and Computers).
@inproceedings{fe8971afb42c42909cb06a4abb4ba8cc,
title = "3D Capsule Networks for Object Classification from 3D Model Data",
abstract = "Many of the existing object classification methods today rely on convolutional neural networks (CNNs), which are very successful in extracting features from the data. However, CNNs cannot sufficiently address the spatial relationship between features and require large amounts of data for training. In this paper, a new architecture is proposed for 3D object classification, which is an extension of the Capsule Networks (CapsNets) to 3D data. Our proposed 3D CapsNet architecture preserves the orientation and spatial relationship of the extracted features, and thus requires less data to train the network. We compare our approach with a ShapeNet inspired model, and show that our method provides performance improvement especially when training data size gets smaller. We also compare and evaluate several different versions of the 3D Capsnet architecture.",
keywords = "3D, Capsule networks, object classification",
author = "Ayesha Ahmad and Burak Kakillioglu and Senem Velipasalar",
year = "2019",
month = "2",
day = "19",
doi = "10.1109/ACSSC.2018.8645256",
language = "English (US)",
series = "Conference Record - Asilomar Conference on Signals, Systems and Computers",
publisher = "IEEE Computer Society",
pages = "2225--2229",
editor = "Matthews, {Michael B.}",
booktitle = "Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018",
address = "United States",

}

TY - GEN

T1 - 3D Capsule Networks for Object Classification from 3D Model Data

AU - Ahmad, Ayesha

AU - Kakillioglu, Burak

AU - Velipasalar, Senem

PY - 2019/2/19

Y1 - 2019/2/19

N2 - Many of the existing object classification methods today rely on convolutional neural networks (CNNs), which are very successful in extracting features from the data. However, CNNs cannot sufficiently address the spatial relationship between features and require large amounts of data for training. In this paper, a new architecture is proposed for 3D object classification, which is an extension of the Capsule Networks (CapsNets) to 3D data. Our proposed 3D CapsNet architecture preserves the orientation and spatial relationship of the extracted features, and thus requires less data to train the network. We compare our approach with a ShapeNet inspired model, and show that our method provides performance improvement especially when training data size gets smaller. We also compare and evaluate several different versions of the 3D Capsnet architecture.

AB - Many of the existing object classification methods today rely on convolutional neural networks (CNNs), which are very successful in extracting features from the data. However, CNNs cannot sufficiently address the spatial relationship between features and require large amounts of data for training. In this paper, a new architecture is proposed for 3D object classification, which is an extension of the Capsule Networks (CapsNets) to 3D data. Our proposed 3D CapsNet architecture preserves the orientation and spatial relationship of the extracted features, and thus requires less data to train the network. We compare our approach with a ShapeNet inspired model, and show that our method provides performance improvement especially when training data size gets smaller. We also compare and evaluate several different versions of the 3D Capsnet architecture.

KW - 3D

KW - Capsule networks

KW - object classification

UR - http://www.scopus.com/inward/record.url?scp=85063006154&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063006154&partnerID=8YFLogxK

U2 - 10.1109/ACSSC.2018.8645256

DO - 10.1109/ACSSC.2018.8645256

M3 - Conference contribution

AN - SCOPUS:85063006154

T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers

SP - 2225

EP - 2229

BT - Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018

A2 - Matthews, Michael B.

PB - IEEE Computer Society

ER -