SIA OpenIR  > 智能产线与系统研究室
Alternative Title3D Shape Classification Based on Deep Belief Network and Spectral Feature
高恩阳; 刘伟军; 王天然; 邓华波
Source Publication机械设计与制造
Contribution Rank1
Funding Organization973项目—数学机械化方法及其在数字化设计制造中的应用(2011CB302400)
Keyword模型分类 深度信任网络 特征提取
Other AbstractIn CAD/CAM,the classification of three-dimensional model is an important issue. Previous work mainly considered the classification of the rigid model. Due to the presence of non-rigid deformation,the difficulty of the three-dimensional model classification problem greatly increased. Three-dimensional model classification method based on spectral feature and depth belief network were proposed. First,spectral feature of a 3D model was extracted; then the main component in the spectral features was extracted by DBN dimensionality reduction;finally,discriminated classification by support vector machine was applied. The experiments showed that the proposed method based on spectral feature and deep belied network can effectively describe the intrinsic feature of the non-rigid three-dimensional model to obtain a better classification results.
Document Type期刊论文
Recommended Citation
GB/T 7714
高恩阳,刘伟军,王天然,等. 基于谱特征和深度信任网络的三维模型分类[J]. 机械设计与制造,2013(3):250-252.
APA 高恩阳,刘伟军,王天然,&邓华波.(2013).基于谱特征和深度信任网络的三维模型分类.机械设计与制造(3),250-252.
MLA 高恩阳,et al."基于谱特征和深度信任网络的三维模型分类".机械设计与制造 .3(2013):250-252.
Files in This Item: Download All
File Name/Size DocType Version Access License
基于谱特征和深度信任网络的三维模型分类.(995KB) 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[高恩阳]'s Articles
[刘伟军]'s Articles
[王天然]'s Articles
Baidu academic
Similar articles in Baidu academic
[高恩阳]'s Articles
[刘伟军]'s Articles
[王天然]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[高恩阳]'s Articles
[刘伟军]'s Articles
[王天然]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 基于谱特征和深度信任网络的三维模型分类.pdf
Format: Adobe PDF
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.