Multi-camera invariant appearance modeling for non-rigid object identification in a real-time environment

NCU Institutional Repository > 工學院 > 機械工程學系 > 期刊論文 > Item 987654321/108971

請使用永久網址來引用或連結此文件: https://ir.lib.ncu.edu.tw/handle/987654321/108971

題名:	Multi-camera invariant appearance modeling for non-rigid object identification in a real-time environment
作者:	林智揚;Lin, Chih-Yang;Kang, Li-Wei;Kao, Jau-Hong;Lu, Chun-Shien;Wu, Yi-Ta
貢獻者:	工學院機械工程學系
關鍵詞:	Appearance model;Gaussian mixture model;Hierarchical tree structures;Identification;Invariant features;Multi-camera;Real-time system;Surveillance
日期:	2013-01-01
上傳時間:	2026-04-23 15:18:29 (UTC+8)
出版者:	Academic Press Inc.;Elsevier Inc
摘要:	摘要： ► Propose a new multi-level and coarse-to-fine structure for a human appearance. ► Combining multiple structures into a unified model by Gaussian mixture models. ► Identify a person using only one frame. ► Proposed model is invariant to rotation, scaling, translation, and shape changes. Surveillance of wide areas requires a system of multiple cameras to keep observing people, the non-rigid objects. In such a multiple view system, the appearance of people obtained in one camera is usually different from the appearance obtained in other cameras. In order to correctly identify people, the unique appearance model of each specific object should be invariant to such changes. Unlike previous methods building an appearance model by using only single camera, our appearance modeling, in this paper, is based on the multi-camera environment to fit real cases. Our appearance model is represented by two hierarchical tree structures that are responsible for color and texture information, respectively, where each layer of a tree is maintained by a Gaussian mixture model (GMM). The identification process is performed with a delicate voting scheme without complicated computations to meet the requirements of real-time applications. Experimental results show that our unique appearance model is robust to translation, rotation, scaling, and shape variations. Furthermore, it is equipped with automatic model updating, and it achieves a high precision rate and high processing performance. 出版者： Elsevier Inc 出版日期： 2013-08-01 出處： Journal of visual communication and image representation, 2013-08, Vol.24 (6), p.717-728 資源來源： Elsevier ScienceDirect Journals Complete 版權： 2012 Elsevier Inc. 識別號： ISSN: 1047-3203 識別號： EISSN: 1095-9076 識別號： DOI: 10.1016/j.jvcir.2012.01.018
顯示於類別:	[機械工程學系] 期刊論文

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	17	檢視/開啟

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....