Robust Multi-Body Motion Segmentation Based on Fuzzy k-Subspace Clustering
Summary :
The problem of multi-body motion segmentation is important in many computer vision applications. In this paper, we propose a novel algorithm called fuzzy k-subspace clustering for robust segmentation. The proposed method exploits the property that under orthographic camera model the tracked feature points of moving objects reside in multiple subspaces. We compute a partition of feature points into corresponding subspace clusters. First, we find a "soft partition" of feature points based on fuzzy k-subspace algorithm. The proposed fuzzy k-subspace algorithm iteratively minimizes the objective function using Weighted Singular Value Decomposition. Then the points with high partition confidence are gathered to form the subspace bases and the remaining points are classified using their distance to the bases. The proposed method can handle the case of missing data naturally, meaning that the feature points do not have to be visible throughout the sequence. The method is robust to noise and insensitive to initialization. Extensive experiments on synthetic and real data show the effectiveness of the proposed fuzzy k-subspace clustering algorithm.
- Publication
- IEICE TRANSACTIONS on Information Vol.E88-D No.11 pp.2609-2614
- Publication Date
- 2005/11/01
- Publicized
- Online ISSN
- DOI
- 10.1093/ietisy/e88-d.11.2609
- Type of Manuscript
- LETTER
- Category
- Image Recognition, Computer Vision
Authors
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Xi LI, Zhengnan NING, Liuwei XIANG, "Robust Multi-Body Motion Segmentation Based on Fuzzy k-Subspace Clustering" in IEICE TRANSACTIONS on Information,
vol. E88-D, no. 11, pp. 2609-2614, November 2005, doi: 10.1093/ietisy/e88-d.11.2609.
Abstract: The problem of multi-body motion segmentation is important in many computer vision applications. In this paper, we propose a novel algorithm called fuzzy k-subspace clustering for robust segmentation. The proposed method exploits the property that under orthographic camera model the tracked feature points of moving objects reside in multiple subspaces. We compute a partition of feature points into corresponding subspace clusters. First, we find a "soft partition" of feature points based on fuzzy k-subspace algorithm. The proposed fuzzy k-subspace algorithm iteratively minimizes the objective function using Weighted Singular Value Decomposition. Then the points with high partition confidence are gathered to form the subspace bases and the remaining points are classified using their distance to the bases. The proposed method can handle the case of missing data naturally, meaning that the feature points do not have to be visible throughout the sequence. The method is robust to noise and insensitive to initialization. Extensive experiments on synthetic and real data show the effectiveness of the proposed fuzzy k-subspace clustering algorithm.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e88-d.11.2609/_p
Copy
@ARTICLE{e88-d_11_2609,
author={Xi LI, Zhengnan NING, Liuwei XIANG, },
journal={IEICE TRANSACTIONS on Information},
title={Robust Multi-Body Motion Segmentation Based on Fuzzy k-Subspace Clustering},
year={2005},
volume={E88-D},
number={11},
pages={2609-2614},
abstract={The problem of multi-body motion segmentation is important in many computer vision applications. In this paper, we propose a novel algorithm called fuzzy k-subspace clustering for robust segmentation. The proposed method exploits the property that under orthographic camera model the tracked feature points of moving objects reside in multiple subspaces. We compute a partition of feature points into corresponding subspace clusters. First, we find a "soft partition" of feature points based on fuzzy k-subspace algorithm. The proposed fuzzy k-subspace algorithm iteratively minimizes the objective function using Weighted Singular Value Decomposition. Then the points with high partition confidence are gathered to form the subspace bases and the remaining points are classified using their distance to the bases. The proposed method can handle the case of missing data naturally, meaning that the feature points do not have to be visible throughout the sequence. The method is robust to noise and insensitive to initialization. Extensive experiments on synthetic and real data show the effectiveness of the proposed fuzzy k-subspace clustering algorithm.},
keywords={},
doi={10.1093/ietisy/e88-d.11.2609},
ISSN={},
month={November},}
Copy
TY - JOUR
TI - Robust Multi-Body Motion Segmentation Based on Fuzzy k-Subspace Clustering
T2 - IEICE TRANSACTIONS on Information
SP - 2609
EP - 2614
AU - Xi LI
AU - Zhengnan NING
AU - Liuwei XIANG
PY - 2005
DO - 10.1093/ietisy/e88-d.11.2609
JO - IEICE TRANSACTIONS on Information
SN -
VL - E88-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2005
AB - The problem of multi-body motion segmentation is important in many computer vision applications. In this paper, we propose a novel algorithm called fuzzy k-subspace clustering for robust segmentation. The proposed method exploits the property that under orthographic camera model the tracked feature points of moving objects reside in multiple subspaces. We compute a partition of feature points into corresponding subspace clusters. First, we find a "soft partition" of feature points based on fuzzy k-subspace algorithm. The proposed fuzzy k-subspace algorithm iteratively minimizes the objective function using Weighted Singular Value Decomposition. Then the points with high partition confidence are gathered to form the subspace bases and the remaining points are classified using their distance to the bases. The proposed method can handle the case of missing data naturally, meaning that the feature points do not have to be visible throughout the sequence. The method is robust to noise and insensitive to initialization. Extensive experiments on synthetic and real data show the effectiveness of the proposed fuzzy k-subspace clustering algorithm.
ER -
English
