- Research Article
13
- 10.1016/j.ins.2017.05.007
Self-regularized fixed-rank representation for subspace segmentation
- May 08, 2017
- Information Sciences
- Lai Wei + 3 more +3
Self-regularized fixed-rank representation for subspace segmentation
Low-rank learning has attracted much attention recently due to its efficacy in a rich variety of real-world tasks, e.g., subspace segmentation and image categorization. Most low-rank methods are incapable of capturing low-dimensional subspace for supervised learning tasks, e.g., classification and regression. This paper aims to learn both the discriminant low-rank representation (LRR) and the robust projecting subspace in a supervised manner. To achieve this goal, we cast the problem into a constrained rank minimization framework by adopting the least squares regularization. Naturally, the data label structure tends to resemble that of the corresponding low-dimensional representation, which is derived from the robust subspace projection of clean data by low-rank learning. Moreover, the low-dimensional representation of original data can be paired with some informative structure by imposing an appropriate constraint, e.g., Laplacian regularizer. Therefore, we propose a novel constrained LRR method. The objective function is formulated as a constrained nuclear norm minimization problem, which can be solved by the inexact augmented Lagrange multiplier algorithm. Extensive experiments on image classification, human pose estimation, and robust face recovery have confirmed the superiority of our method.
Self-regularized fixed-rank representation for subspace segmentation
Self-regularized fixed-rank representation for subspace segmentation
Low-rank matrix decomposition method for potential field data separation
Separation of potential field data forms the basis of inversion and interpretation. The low-rank matrix theory is used for the separation of potential field data. A theoretical analysis led to the approximate equation that demonstrates the relation between the amplitudes of the wavenumber components of potential field data and the singular values of the trajectory matrix embedded from the potential field data matrix. Therefore, the low-rank feature of the trajectory matrix of regional field data and the sparse feature of the trajectory matrix of residual field data can be obtained based on the features of the power spectrum of the potential field data. Based on this, we have developed a low-rank matrix decomposition model for the separation of the trajectory matrix of the potential field data. Minimizing the rank of the trajectory matrix of the regional field data and the ℓ0-norm of the trajectory matrix of the residual field data is a double-objective optimization task, and this optimization task can be solved by the inexact augmented Lagrange multiplier algorithm. The weighting parameter is robust and easy to set. Numerical experiment results indicate that our method is robust, and the separation errors of the method are smaller compared to the matched filtering and wavelet analysis methods. In addition, our method can be easily implemented and has clear geophysical significance. Finally, our method is applied on real data sets in the Daye area, Hubei Province, China. The separated gravity and magnetic fields coincide well with target geologic sources.
Read moreCorrelation Adaptive Subspace Segmentation by Trace Lasso
This paper studies the subspace segmentation problem. Given a set of data points drawn from a union of subspaces, the goal is to partition them into their underlying subspaces they were drawn from. The spectral clustering method is used as the framework. It requires to find an affinity matrix which is close to block diagonal, with nonzero entries corresponding to the data point pairs from the same subspace. In this work, we argue that both sparsity and the grouping effect are important for subspace segmentation. A sparse affinity matrix tends to be block diagonal, with less connections between data points from different subspaces. The grouping effect ensures that the highly corrected data which are usually from the same subspace can be grouped together. Sparse Subspace Clustering (SSC), by using $\ell^1$-minimization, encourages sparsity for data selection, but it lacks of the grouping effect. On the contrary, Low-Rank Representation (LRR), by rank minimization, and Least Squares Regression (LSR), by $\ell^2$-regularization, exhibit strong grouping effect, but they are short in subset selection. Thus the obtained affinity matrix is usually very sparse by SSC, yet very dense by LRR and LSR. In this work, we propose the Correlation Adaptive Subspace Segmentation (CASS) method by using trace Lasso. CASS is a data correlation dependent method which simultaneously performs automatic data selection and groups correlated data together. It can be regarded as a method which adaptively balances SSC and LSR. Both theoretical and experimental results show the effectiveness of CASS.
Read moreOn the Robustness of 3D Human Pose Estimation
It is widely shown that Convolutional Neural Networks (CNNs) are vulnerable to adversarial examples on most recognition tasks, such as image classification and segmentation. However, few work studies the more complicated task – 3D human pose estimation. This task often requires large-scale datasets, specialized network architectures, and it can be solved either from single-view RGB images or from multi-view RGB images. In this paper, we make the first attempt to investigate the robustness of current state-of-the-art 3D human pose estimation methods. To this end, we build four representative baseline models, where most of the current methods can be generally classified as one of them. Furthermore, we design targeted adversarial attacks to detect whether 3D pose estimators are robust to different camera parameters. For different types of methods, we present a comprehensive study of their robustness on the large-scale Human3.6M benchmark. Our work shows that different methods vary significantly in their resistance to adversarial attacks. Through extensive experiments, we show that multi-view 3D pose estimators can be more vulnerable to adversarial examples. We believe that our efforts can shed light on future works to design more robust 3D human pose estimators.
Read moreSubspace segmentation via self-regularized latent K-means
Subspace segmentation via self-regularized latent K-means
Unsupervised dimensionality reduction by jointing dynamic hypergraph and low-rank embedding for classification and clustering
Unsupervised dimensionality reduction by jointing dynamic hypergraph and low-rank embedding for classification and clustering
Read moreA regularized low-rank representation model for facial expression recognition
Subspace learning plays a key role in pattern recognition and machine learning. However, its performance would be degraded when data are corrupted by various occlusions. Low-rank representation (LRR) can recover the corrupted data and explore low-dimensional subspace structures embedded in data. Inspired by low-rank representation and subspace learning, in this paper, we propose a regularized low-rank representation model, which not only learns a low-rank matrix that can capture the global structure of data, but also derives a robust subspace from the corrupted data. The problem can be formulated a rank minimization objective function, which is effectively solved by numerical algorithm. The extensive experimental results on two facial expression databases demonstrate the effectiveness of the proposed method.
Read moreDownlink channel estimation for millimeter wave communication combining low-rank and sparse structure characteristics
The acquisition of channel state information (CSI) is essential in millimeter wave (mmWave) multiple-input multiple-output (MIMO) systems. The mmWave channel exhibits sparse scattering characteristics and a meaningful low-rank structure, which can be simultaneously employed to reduce the complexity of channel estimation. Most existing works recover the low-rank structure of channels using nuclear norm theory. However, solving the nuclear norm-based convex problem often leads to a suboptimal solution of the rank minimization problem, thus degrading the accuracy of channel estimation. Previous contributions recover the channel using over-complete dictionary with the assumption that the mmWave channel can be sparsely represented under some dictionary. While over-complete dictionary may increase the computational complexity. To address these problems, we propose a channel estimation framework based on non-convex low-rank approximation and dictionary learning by exploring the joint low-rank and sparse representations of wireless channels. We surrogate the widely used nuclear norm theory with non-convex low-rank approximation method and design a dictionary learning algorithm based on channel feature classification employing deep neural network (DNN). Our simulation results reveal the proposed scheme outperform the conventional dictionary learning algorithm, Bayesian framework algorithm, and compressed sensing-based algorithms.
Read moreLearning Tensor Low-Rank Representation for Hyperspectral Anomaly Detection.
Recently, low-rank representation (LRR) methods have been widely applied for hyperspectral anomaly detection, due to their potentials in separating the backgrounds and anomalies. However, existing LRR models generally convert 3-D hyperspectral images (HSIs) into 2-D matrices, inevitably leading to the destruction of intrinsic 3-D structure properties in HSIs. To this end, we propose a novel tensor low-rank and sparse representation (TLRSR) method for hyperspectral anomaly detection. A 3-D TLR model is expanded to separate the LR background part represented by a tensorial background dictionary and corresponding coefficients. This representation characterizes the multiple subspace property of the complex LR background. Based on the weighted tensor nuclear norm and the LF,1 sparse norm, a dictionary is designed to make its atoms more relevant to the background. Moreover, a principal component analysis (PCA) method can be assigned as one preprocessing step to exact a subset of HSI bands, retaining enough the HSI object information and reducing computational time of the postprocessing tensorial operations. The proposed model is efficiently solved by the well-designed alternating direction method of multipliers (ADMMs). A comparison with the existing algorithms via experiments establishes the competitiveness of the proposed method with the state-of-the-art competitors in the hyperspectral anomaly detection task.
Read moreOn Classification of the Human Emotions from Facial Thermal Images: A Case Study Based on Machine Learning
(1) Background: This paper intends to accomplish a comparative study and analysis regarding the multiclass classification of facial thermal images, i.e., in three classes corresponding to predefined emotional states (neutral, happy and sad). By carrying out a comparative analysis, the main goal of the paper consists in identifying a suitable algorithm from machine learning field, which has the highest accuracy (ACC). Two categories of images were used in the process, i.e., images with Gaussian noise and images with “salt and pepper” type noise that come from two built-in special databases. An augmentation process was applied to the initial raw images that led to the development of the two databases with added noise, as well as the subsequent augmentation of all images, i.e., rotation, reflection, translation and scaling. (2) Methods: The multiclass classification process was implemented through two subsets of methods, i.e., machine learning with random forest (RF), support vector machines (SVM) and k-nearest neighbor (KNN) algorithms and deep learning with the convolutional neural network (CNN) algorithm. (3) Results: The results obtained in this paper with the two subsets of methods belonging to the field of artificial intelligence (AI), together with the two categories of facial thermal images with added noise used as input, were very good, showing a classification accuracy of over 99% for the two categories of images, and the three corresponding classes for each. (4) Discussion: The augmented databases and the additional configurations of the implemented algorithms seems to have had a positive effect on the final classification results.
Read moreLaplacian regularized locality-constrained coding for image classification
Laplacian regularized locality-constrained coding for image classification
A Two-Stream Unified Interpretation Network for Heterogeneous Remote Sensing Images Classification
The conventional studies on different types of remote sensing (RS) images classifications are conducted separately. Thanks to the powerful potential of deep learning to automatically learn features from data, exploring a unified method is possible. Moreover, recent research shows that sparse and low-rank representations can convey valuable information for patterns classification. Therefore, this paper presents a two-stream heterogeneous RS images unified interpretation network (HRSIUI-Net). One stream is to transfer the pre-trained fully convolutional network to learn deep multi-scale spatial features of RS data. The other stream is to employ a subspace learning based on graph embedding to learn the sparse and low-rank subspace representations of high-dimensional features. And then, two streams of learned subspace features are integrated for classification combined with an SVM. The experimental results on two typical RS data indicate that HRSIUI-Net can achieve competitive performance.
Read moreMulti-cancer samples clustering via graph regularized low-rank representation method under sparse and symmetric constraints
BackgroundIdentifying different types of cancer based on gene expression data has become hotspot in bioinformatics research. Clustering cancer gene expression data from multiple cancers to their own class is a significance solution. However, the characteristics of high-dimensional and small samples of gene expression data and the noise of the data make data mining and research difficult. Although there are many effective and feasible methods to deal with this problem, the possibility remains that these methods are flawed.ResultsIn this paper, we propose the graph regularized low-rank representation under symmetric and sparse constraints (sgLRR) method in which we introduce graph regularization based on manifold learning and symmetric sparse constraints into the traditional low-rank representation (LRR). For the sgLRR method, by means of symmetric constraint and sparse constraint, the effect of raw data noise on low-rank representation is alleviated. Further, sgLRR method preserves the important intrinsic local geometrical structures of the raw data by introducing graph regularization. We apply this method to cluster multi-cancer samples based on gene expression data, which improves the clustering quality. First, the gene expression data are decomposed by sgLRR method. And, a lowest rank representation matrix is obtained, which is symmetric and sparse. Then, an affinity matrix is constructed to perform the multi-cancer sample clustering by using a spectral clustering algorithm, i.e., normalized cuts (Ncuts). Finally, the multi-cancer samples clustering is completed.ConclusionsA series of comparative experiments demonstrate that the sgLRR method based on low rank representation has a great advantage and remarkable performance in the clustering of multi-cancer samples.
Read moreLow-rank inter-class sparsity based semi-flexible target least squares regression for feature representation
Low-rank inter-class sparsity based semi-flexible target least squares regression for feature representation
Subspace Segmentation and Clustering of Medical Image Data Based on Low-Rank Representation
Biomedical cell image contains different types of components. The segmentation of these components is essential for characterizing the overall structure of leucocytes. However, the task is challenging because the data generally possess irregular patterns, varying eccentricity and congestion of inter class substructures. In order to perform the automatic segmentation of leucocytes, many approaches in machine learning have been proposed. Most of the recent methods utilize k-means and fuzzy c-means algorithms. These methods suffer from the under-segmentation and the oversegmentation of data points. We have adopted an efficient Low Rank Representation (LRR) method for the segmentation of cell image patterns. Its overall segmentation performance is very high in comparison to the k-means and fuzzy c-means segmentation methods. It discriminates sub structures by incorporating multiple feature data into the joint representation matrix. The main improvements of our adopted approach, include the global geometric representation of different cell patterns and the control of different sources of noise in the data. Data points of the cell image are evolved with the basic assumption as x = m + n. Where x is the input observation, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$m$</tex> is the set of data points that lies in the correct subspace and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$n$</tex> represent the noise in the data. The performance of the model was evaluated on the different types of medical images. It has reported the clustering accuracy of 97.98% on the biomedical database. Whereas an accuracy of 93.1 % was reported on ALL-IDB cell dataset.
Read more