Skip to main navigation Skip to search Skip to main content

High-level semantic feature for 3D shape based on deep belief networks

  • Northwestern Polytechnical University Xian
  • Information Engineering University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

17 Scopus citations

Abstract

Deep learning has emerged as a powerful technique to extract high-level features from low-level information, which shows that hierarchical representation can be easily achieved. However, applying deep learning into 3D shape is still a challenge. In this paper, we propose a novel high-level feature learning method for 3D shape retrieval based on deep learning. In this framework, the low-level 3D shape descriptors are first encoded into visual bag-of-words, and then highlevel shape features are generated via deep belief network, which facilitates a good semantic preserving ability for the tasks of shape classification and retrieval. Experiments on 3D shape recognition and retrieval demonstrate the superior performance of the proposed method in comparison to the state-of-the-art methods.

Original languageEnglish
Title of host publication2014 IEEE International Conference on Multimedia and Expo, ICME 2014
PublisherIEEE Computer Society
EditionSeptmber
ISBN (Electronic)9781479947614
DOIs
StatePublished - 3 Sep 2014
Event2014 IEEE International Conference on Multimedia and Expo, ICME 2014 - Chengdu, China
Duration: 14 Jul 201418 Jul 2014

Publication series

NameProceedings - IEEE International Conference on Multimedia and Expo
NumberSeptmber
Volume2014-September
ISSN (Print)1945-7871
ISSN (Electronic)1945-788X

Conference

Conference2014 IEEE International Conference on Multimedia and Expo, ICME 2014
Country/TerritoryChina
CityChengdu
Period14/07/1418/07/14

Keywords

  • 3D Shape classification
  • 3D shape retrieval
  • Bag-of-words
  • Deep belief networks
  • Deep learning

Fingerprint

Dive into the research topics of 'High-level semantic feature for 3D shape based on deep belief networks'. Together they form a unique fingerprint.

Cite this