3D model retrieval by sample based alignment

NCUIR > School of Management at National Central University > Department of Information Management > journal & Dissertation > Item 987654321/106170

Please use this identifier to cite or link to this item: https://ir.lib.ncu.edu.tw/handle/987654321/106170

Title:	3D model retrieval by sample based alignment
Authors:	蔡志豐;Chen, Zong-Yao;Lin, Wei-Chao;Tsai, Chih-Fong;Ke, Shih-Wen
Contributors:	管理學院資訊管理學系
Keywords:	3D model;Continuous Principal Component Analysis (CPCA);LightField Descriptors (LFD);Multimedia;Retrieval
Date:	2016-10-01
Issue Date:	2026-04-23 13:11:52 (UTC+8)
Publisher:	Academic Press Inc.;Elsevier Inc
Abstract:	摘要： •The Sample Based Alignment (SBA) method is proposed for better 3D model alignment and retrieval.•The SBA method can provide the aligned model angle and boost retrieval performance.•The experiments are based on the Engineering Shape Benchmark (ESB) and Princeton Shape Benchmark (PSB) datasets.•SBA outperforms LFD, CPCA and other baselines for 3D model retrieval and classification. In 3D model retrieval, preprocessing of 3D models is needed, in which alignment is a key factor that significantly affects retrieval performance. In particular, the anti-rotation image feature can obtain the alignment effect of 3D model views. In practice, the focus of many users of 3D models is not just on retrieval performance, but the use of aligned models for different purposes. In this paper, we propose a method, namely Sample Based Alignment (SBA) for better 3D model alignment and retrieval. In SBA, given a class, a sample model is used as the target for alignment, after which each 3D model in this class is then aligned one by one, i.e., the 3D model is actually rotated. Our experimental results, based on two 3D model datasets and performance comparisons with other methods, demonstrate the superiority of the SBA method over state-of-the-art methods in terms of 3D model retrieval and classification. 出版者： Elsevier Inc 出版日期： 2016-10-01 出處： Journal of visual communication and image representation, 2016-10, Vol.40, p.721-731 資源來源： Elsevier ScienceDirect Journals Complete 版權： 2016 Elsevier Inc. 識別號： ISSN: 1047-3203 識別號： EISSN: 1095-9076 識別號： DOI: 10.1016/j.jvcir.2016.08.017
Appears in Collections:	[Department of Information Management] journal & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	36	View/Open

社群 sharing

Loading...