Face clustering via learning a sparsity preserving low-rank graph
- PDF / 1,833,637 Bytes
- 20 Pages / 439.642 x 666.49 pts Page_size
- 88 Downloads / 186 Views
Face clustering via learning a sparsity preserving low-rank graph Changpeng Wang1 · Jiangshe Zhang2 · Xueli Song1 · Tianjun Wu1 Received: 12 October 2019 / Revised: 22 May 2020 / Accepted: 21 July 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Face clustering aims to group the face images without any label information into clusters, and has recently attracted considerable attention in machine learning and data mining. Many graph based clustering methods have been proposed and among which sparse representation (SR) and low-rank representation (LRR) are two representative methods for affinity graph construction. The clustering result may be inaccurate if the affinity graph is constructed with low quality. In this paper, we propose a novel face clustering method via learning a sparsity preserving low-rank graph (LSPLRG), where the initial affinity graph is derived on the sparse coefficients without any a priori graph or similarity matrix. In addition, an adaptive weighted matrix is imposed on the data reconstruction errors to enhance the role of important features, while a constraint on the representation matrix is to reduce the redundant features. By integrating the local distance regularization term into LRR, LSPLRG could exploit the global and local structures of data simultaneously. These appealing properties allow LSPLRG to well capture the intrinsic structure of data, and thus has potential to improve clustering performance. Experiments conducted on several face image databases demonstrate the effectiveness and robustness of LSPLRG compared with several state-ofthe-art subspace clustering methods. Keywords Low-rank representation · Graph learning · Face clustering
1 Introduction Clustering is a fundamentally important task to numerous applications, such as image classification [34], saliency detection [40], image segmentation [44] and motion segmentation [31]. The goal of clustering is to simultaneously segment unlabeled data points into clusters so that the data points in the same clusters are more similar to each other than those
Xueli Song
[email protected] 1
School of Science, Chang’an University, Xi’an, 710064, China
2
School of Mathematics and Statistics, Xi’an Jiaotong University, Xi’an, 710049, China
Multimedia Tools and Applications
in different clusters [38]. Under the Lambertian assumption, the face images of a subject with a fixed pose and varying illumination approximately lie in a linear subspace of dimension 9 [1]. Thus, the face clustering problem can be considered as image clustering problem over a union of subspaces. In the past decades, a large number of clustering methods have emerged, such as k-means [26], spectral clustering [28], support vector clustering [2], maximum margin clustering [37] and multi-view subspace clustering [42, 48, 49]. In the big data era, high-dimensional data is ubiquitous in many real applications such as image processing [18, 20]. However, high-dimensional data not only results in high computational cost of time and me
Data Loading...
