CASIA OpenIR  > 模式识别国家重点实验室  > 图像与视频分析
Multi-View 3D Object Retrieval With Deep Embedding Network
Guo, Haiyun1,2; Wang, Jinqiao1,2; Gao, Yue3; Li, Jianqiang4; Lu, Hanqing1,2
AbstractIn multi-view 3D object retrieval, each object is characterized by a group of 2D images captured from different views. Rather than using hand-crafted features, in this paper, we take advantage of the strong discriminative power of convolutional neural network to learn an effective 3D object representation tailored for this retrieval task. Specifically, we propose a deep embedding network jointly supervised by classification loss and triplet loss to map the high-dimensional image space into a low-dimensional feature space, where the Euclidean distance of features directly corresponds to the semantic similarity of images. By effectively reducing the intra-class variations while increasing the inter-class ones of the input images, the network guarantees that similar images are closer than dissimilar ones in the learned feature space. Besides, we investigate the effectiveness of deep features extracted from different layers of the embedding network extensively and find that an efficient 3D object representation should be a tradeoff between global semantic information and discriminative local characteristics. Then, with the set of deep features extracted from different views, we can generate a comprehensive description for each 3D object and formulate the multi-view 3D object retrieval as a set-to-set matching problem. Extensive experiments on SHREC'15 data set demonstrate the superiority of our proposed method over the previous state-of-the-art approaches with over 12% performance improvement.
KeywordConvolutional Neural Network Multi-view 3d Object Retrieval Triplet Loss
WOS HeadingsScience & Technology ; Technology
Indexed BySCI
Funding Organization863 Program(2014AA015104) ; National Natural Science Foundation of China(61273034 ; 61332016)
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS IDWOS:000388205100002
Citation statistics
Cited Times:26[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Affiliation1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China
3.Tsinghua Univ, Sch Software, Tsinghua Natl Lab Informat Sci & Technol TNList, Key Lab Informat Syst Secur,Minist Educ, Beijing 100084, Peoples R China
4.Beijing Univ Technol, Sch Software Engn, Beijing Engn Res Ctr IoT Software & Syst, Beijing 100083, Peoples R China
First Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Guo, Haiyun,Wang, Jinqiao,Gao, Yue,et al. Multi-View 3D Object Retrieval With Deep Embedding Network[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2016,25(12):5526-5537.
APA Guo, Haiyun,Wang, Jinqiao,Gao, Yue,Li, Jianqiang,&Lu, Hanqing.(2016).Multi-View 3D Object Retrieval With Deep Embedding Network.IEEE TRANSACTIONS ON IMAGE PROCESSING,25(12),5526-5537.
MLA Guo, Haiyun,et al."Multi-View 3D Object Retrieval With Deep Embedding Network".IEEE TRANSACTIONS ON IMAGE PROCESSING 25.12(2016):5526-5537.
Files in This Item: Download All
File Name/Size DocType Version Access License
HaiyunGuo-multi-view(2793KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Guo, Haiyun]'s Articles
[Wang, Jinqiao]'s Articles
[Gao, Yue]'s Articles
Baidu academic
Similar articles in Baidu academic
[Guo, Haiyun]'s Articles
[Wang, Jinqiao]'s Articles
[Gao, Yue]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Guo, Haiyun]'s Articles
[Wang, Jinqiao]'s Articles
[Gao, Yue]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: HaiyunGuo-multi-view 3D object retrieval with deep embedding network.pdf
Format: Adobe PDF
This file does not support browsing at this time
All comments (0)
No comment.

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