Publications by authors named "Junbin Gao"

53 Publications

Rank Consistency Induced Multiview Subspace Clustering via Low-Rank Matrix Factorization.

IEEE Trans Neural Netw Learn Syst 2021 Apr 21;PP. Epub 2021 Apr 21.

Multiview subspace clustering has been demonstrated to achieve excellent performance in practice by exploiting multiview complementary information. One of the strategies used in most existing methods is to learn a shared self-expressiveness coefficient matrix for all the view data. Different from such a strategy, this article proposes a rank consistency induced multiview subspace clustering model to pursue a consistent low-rank structure among view-specific self-expressiveness coefficient matrices. To facilitate a practical model, we parameterize the low-rank structure on all self-expressiveness coefficient matrices through the tri-factorization along with orthogonal constraints. This specification ensures that self-expressiveness coefficient matrices of different views have the same rank to effectively promote structural consistency across multiviews. Such a model can learn a consistent subspace structure and fully exploit the complementary information from the view-specific self-expressiveness coefficient matrices, simultaneously. The proposed model is formulated as a nonconvex optimization problem. An efficient optimization algorithm with guaranteed convergence under mild conditions is proposed. Extensive experiments on several benchmark databases demonstrate the advantage of the proposed model over the state-of-the-art multiview clustering approaches.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2021.3071797DOI Listing
April 2021

A Decoder-Free Variational Deep Embedding for Unsupervised Clustering.

IEEE Trans Neural Netw Learn Syst 2021 Apr 21;PP. Epub 2021 Apr 21.

In deep clustering frameworks, autoencoder (AE)- or variational AE-based clustering approaches are the most popular and competitive ones that encourage the model to obtain suitable representations and avoid the tendency for degenerate solutions simultaneously. However, for the clustering task, the decoder for reconstructing the original input is usually useless when the model is finished training. The encoder-decoder architecture limits the depth of the encoder so that the learning capacity is reduced severely. In this article, we propose a decoder-free variational deep embedding for unsupervised clustering (DFVC). It is well known that minimizing reconstruction error amounts to maximizing a lower bound on the mutual information (MI) between the input and its representation. That provides a theoretical guarantee for us to discard the bloated decoder. Inspired by contrastive self-supervised learning, we can directly calculate or estimate the MI of the continuous variables. Specifically, we investigate unsupervised representation learning by simultaneously considering the MI estimation of continuous representations and the MI computation of categorical representations. By introducing the data augmentation technique, we incorporate the original input, the augmented input, and their high-level representations into the MI estimation framework to learn more discriminative representations. Instead of matching to a simple standard normal distribution adversarially, we use end-to-end learning to constrain the latent space to be cluster-friendly by applying the Gaussian mixture distribution as the prior. Extensive experiments on challenging data sets show that our model achieves higher performance over a wide range of state-of-the-art clustering approaches.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2021.3071275DOI Listing
April 2021

Artificial intelligence (AI) for medical imaging to combat coronavirus disease (COVID-19): a detailed review with direction for future research.

Artif Intell Rev 2021 Apr 15:1-31. Epub 2021 Apr 15.

Discipline of Business Analytics in Business School, The University of Sydney, Sydney, Australia.

Since early 2020, the whole world has been facing the deadly and highly contagious disease named coronavirus disease (COVID-19) and the World Health Organization declared the pandemic on 11 March 2020. Over 23 million positive cases of COVID-19 have been reported till late August 2020. Medical images such as chest X-rays and Computed Tomography scans are becoming one of the main leading clinical diagnosis tools in fighting against COVID-19, underpinned by Artificial Intelligence based techniques, resulting in rapid decision-making in saving lives. This article provides an extensive review of AI-based methods to assist medical practitioners with comprehensive knowledge of the efficient AI-based methods for efficient COVID-19 diagnosis. Nearly all the reported methods so far along with their pros and cons as well as recommendations for improvements are discussed, including image acquisition, segmentation, classification, and follow-up diagnosis phases developed between 2019 and 2020. AI and machine learning technologies have boosted the accuracy of Covid-19 diagnosis, and most of the widely used deep learning methods have been implemented and worked well with a small amount of data for COVID-19 diagnosis. This review presents a detailed mythological analysis for the evaluation of AI-based methods used in the process of detecting COVID-19 from medical images. However, due to the quick outbreak of Covid-19, there are not many ground-truth datasets available for the communities. It is necessary to combine clinical experts' observations and information from images to have a reliable and efficient COVID-19 diagnosis. This paper suggests that future research may focus on multi-modality based models as well as how to select the best model architecture where AI can introduce more intelligence to medical systems to capture the characteristics of diseases by learning from multi-modality data to obtain reliable results for COVID-19 diagnosis for timely treatment .
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s10462-021-09985-zDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8047522PMC
April 2021

Control the Neural Stem Cell Fate with Biohybrid Piezoelectrical Magnetite Micromotors.

Nano Lett 2021 Apr 13;21(8):3518-3526. Epub 2021 Apr 13.

School of Materials Science and Engineering, Sun Yat-Sen University, Guangzhou 510275, China.

Inducing neural stem cells to differentiate and replace degenerated functional neurons represents the most promising approach for neural degenerative diseases including Parkinson's disease, Alzheimer's disease, etc. While diverse strategies have been proposed in recent years, most of these are hindered due to uncontrollable cell fate and device invasiveness. Here, we report a minimally invasive micromotor platform with biodegradable helical () as the framework and superparamagnetic FeO nanoparticles/piezoelectric BaTiO nanoparticles as the built-in function units. With a low-strength rotational magnetic field, this integrated micromotor system can perform precise navigation in biofluid and achieve single-neural stem cell targeting. Remarkably, by tuning ultrasound intensity, thus the local electrical output by the motor, directed differentiation of the neural stem cell into astrocytes, functional neurons (dopamine neurons, cholinergic neurons), and oligodendrocytes, can be achieved. This micromotor platform can serve as a highly controllable wireless tool for bioelectronics and neuronal regenerative therapy.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acs.nanolett.1c00290DOI Listing
April 2021

Short-Term Lesion Change Detection for Melanoma Screening With Novel Siamese Neural Network.

IEEE Trans Med Imaging 2021 Mar 2;40(3):840-851. Epub 2021 Mar 2.

Short-term monitoring of lesion changes has been a widely accepted clinical guideline for melanoma screening. When there is a significant change of a melanocytic lesion at three months, the lesion will be excised to exclude melanoma. However, the decision on change or no-change heavily depends on the experience and bias of individual clinicians, which is subjective. For the first time, a novel deep learning based method is developed in this paper for automatically detecting short-term lesion changes in melanoma screening. The lesion change detection is formulated as a task measuring the similarity between two dermoscopy images taken for a lesion in a short time-frame, and a novel Siamese structure based deep network is proposed to produce the decision: changed (i.e. not similar) or unchanged (i.e. similar enough). Under the Siamese framework, a novel structure, namely Tensorial Regression Process, is proposed to extract the global features of lesion images, in addition to deep convolutional features. In order to mimic the decision-making process of clinicians who often focus more on regions with specific patterns when comparing a pair of lesion images, a segmentation loss (SegLoss) is further devised and incorporated into the proposed network as a regularization term. To evaluate the proposed method, an in-house dataset with 1,000 pairs of lesion images taken in a short time-frame at a clinical melanoma centre was established. Experimental results on this first-of-a-kind large dataset indicate that the proposed model is promising in detecting the short-term lesion change for objective melanoma screening.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2020.3037761DOI Listing
March 2021

Adaptive Fusion of Heterogeneous Manifolds for Subspace Clustering.

IEEE Trans Neural Netw Learn Syst 2020 Aug 10;PP. Epub 2020 Aug 10.

Multiview clustering (MVC) has recently received great interest due to its pleasing efficacy in combining the abundant and complementary information to improve clustering performance, which overcomes the drawbacks of view limitation existed in the standard single-view clustering. However, the existing MVC methods are mostly designed for vectorial data from linear spaces and, thus, are not suitable for multiple dimensional data with intrinsic nonlinear manifold structures, e.g., videos or image sets. Some works have introduced manifolds' representation methods of data into MVC and obtained considerable improvements, but how to fuse multiple manifolds efficiently for clustering is still a challenging problem. Particularly, for heterogeneous manifolds, it is an entirely new problem. In this article, we propose to represent the complicated multiviews' data as heterogeneous manifolds and a fusion framework of heterogeneous manifolds for clustering. Different from the empirical weighting methods, an adaptive fusion strategy is designed to weight the importance of different manifolds in a data-driven manner. In addition, the low-rank representation is generalized onto the fused heterogeneous manifolds to explore the low-dimensional subspace structures embedded in data for clustering. We assessed the proposed method on several public data sets, including human action video, facial image, and traffic scenario video. The experimental results show that our method obviously outperforms a number of state-of-the-art clustering methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2020.3011717DOI Listing
August 2020

Bone-Targeting Prodrug Mesoporous Silica-Based Nanoreactor with Reactive Oxygen Species Burst for Enhanced Chemotherapy.

ACS Appl Mater Interfaces 2020 Aug 21;12(31):34630-34642. Epub 2020 Jul 21.

School of Pharmaceutical Sciences, Guangdong Provincial Key Laboratory of New Drug Screening, Southern Medical University, Guangzhou 510515, China.

Cancer remains a primary threat to human lives. Recently, amplification of tumor-associated reactive oxygen species (ROS) has been used as a boosting strategy to improve tumor therapy. Here, we report on a bone-targeting prodrug mesoporous silica-based nanoreactor for combined photodynamic therapy (PDT) and enhanced chemotherapy for osteosarcoma. Because of surface modification of a bone-targeting biphosphate moiety and the enhanced permeability and retention effect, the formed nanoreactor shows efficient accumulation in osteosarcoma and exhibits long-term retention in the tumor microenvironment. Upon laser irradiation, the loaded photosensitizer chlorin e6 (Ce6) produces in situ ROS, which not only works for PDT but also functions as a trigger for controlled release of doxorubicin (DOX) and doxycycline (DOXY) from the prodrugs based on a thioketal () linkage. The released DOXY further promotes ROS production, thus perpetuating subsequent DOX/DOXY release and ROS burst. The ROS amplification induces long-term high oxidative stress, which increases the sensitivity of the osteosarcoma to chemotherapy, therefore resulting in enhanced tumor cell inhibition and apoptosis. The as-developed nanoreactor with combined PDT and enhanced chemotherapy based on ROS amplification shows significant promise as a potential platform for cancer treatment.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acsami.0c08992DOI Listing
August 2020

Hyperthermia-Triggered On-Demand Biomimetic Nanocarriers for Synergetic Photothermal and Chemotherapy.

Adv Sci (Weinh) 2020 Jun 20;7(11):1903642. Epub 2020 Apr 20.

School of Pharmaceutical Science Guangdong Provincial Key Laboratory of New Drug Screening Southern Medical University Guangzhou 510515 China.

Nanoparticle-based drug delivery systems with low side effects and enhanced efficacy hold great potential in the treatment of various malignancies, in particular cancer; however, they are still challenging to attain. Herein, an anticancer drug delivery system based on a cisplatin (CDDP) containing nanogel, functionalized with photothermal gold nanorods (GNRs) which are electrostatically decorated with doxorubicin (DOX) is reported. The nanoparticles are formed via the crosslinking reaction of hyaluronic acid with the ancillary anticarcinogen CDDP in the presence of DOX-decorated GNRs. The nanogel is furthermore cloaked with a cancer cell membrane, and the resulting biomimetic nanocarrier (4T1-HANG-GNR-DC) shows efficient accumulation by homologous tumor targeting and possesses long-time retention in the tumor microenvironment. Upon near-infrared (NIR) laser irradiation, in situ photothermal therapy is conducted which further induces hyperthermia-triggered on-demand drug release from the nanogel reservoir to achieve a synergistic photothermal/chemo-therapy. The as-developed biomimetic nanocarriers, with their dual-drug delivery features, homotypic tumor targeting and synergetic photothermal/chemo-therapy, show much promise as a potential platform for cancer treatment.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/advs.201903642DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7284223PMC
June 2020

Probabilistic Linear Discriminant Analysis Based on L₁-Norm and Its Bayesian Variational Inference.

IEEE Trans Cybern 2020 May 5. Epub 2020 May 5.

Probabilistic linear discriminant analysis (PLDA) is a very effective feature extraction approach and has obtained extensive and successful applications in supervised learning tasks. It employs the squared L₂-norm to measure the model errors, which assumes a Gaussian noise distribution implicitly. However, the noise in real-life applications may not follow a Gaussian distribution. Particularly, the squared L₂-norm could extremely exaggerate data outliers. To address this issue, this article proposes a robust PLDA model under the assumption of a Laplacian noise distribution, called L1-PLDA. The learning process employs the approach by expressing the Laplacian density function as a superposition of an infinite number of Gaussian distributions via introducing a new latent variable and then adopts the variational expectation-maximization (EM) algorithm to learn parameters. The most significant advantage of the new model is that the introduced latent variable can be used to detect data outliers. The experiments on several public databases show the superiority of the proposed L1-PLDA model in terms of classification and outlier detection.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TCYB.2020.2985997DOI Listing
May 2020

Robust Functional Manifold Clustering.

IEEE Trans Neural Netw Learn Syst 2021 Feb 4;32(2):777-787. Epub 2021 Feb 4.

In machine learning, it is common to interpret each data sample as a multivariate vector disregarding the correlations among covariates. However, the data may actually be functional, i.e., each data point is a function of some variable, such as time, and the function is discretely sampled. The naive treatment of functional data as traditional multivariate data can lead to poor performance due to the correlations. In this article, we focus on subspace clustering for functional data or curves and propose a new method robust to shift and rotation. The idea is to define a function or curve and all its versions generated by shift and rotation as an equivalent class and then to find the subspace structure among all equivalent classes as the surrogate for all curves. Experimental evaluation on synthetic and real data reveals that this method massively outperforms prior clustering methods in both speed and accuracy when clustering functional data.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2020.2979444DOI Listing
February 2021

Deep Coattention-Based Comparator for Relative Representation Learning in Person Re-Identification.

IEEE Trans Neural Netw Learn Syst 2021 Feb 4;32(2):722-735. Epub 2021 Feb 4.

Person re-identification (re-ID) favors discriminative representations over unseen shots to recognize identities in disjoint camera views. Effective methods are developed via pair-wise similarity learning to detect a fixed set of region features, which can be mapped to compute the similarity value. However, relevant parts of each image are detected independently without referring to the correlation on the other image. Also, region-based methods spatially position local features for their aligned similarities. In this article, we introduce the deep coattention-based comparator (DCC) to fuse codependent representations of paired images so as to correlate the best relevant parts and produce their relative representations accordingly. The proposed approach mimics the human foveation to detect the distinct regions concurrently across images and alternatively attends to fuse them into the similarity learning. Our comparator is capable of learning representations relative to a test shot and well-suited to reidentifying pedestrians in surveillance. We perform extensive experiments to provide the insights and demonstrate the state of the arts achieved by our method in benchmark data sets: 1.2 and 2.5 points gain in mean average precision (mAP) on DukeMTMC-reID and Market-1501, respectively.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2020.2979190DOI Listing
February 2021

Branched convolutional neural networks incorporated with Jacobian deep regression for facial landmark detection.

Neural Netw 2019 Oct 19;118:127-139. Epub 2019 Jun 19.

The University of Sydney Business School, The University of Sydney, Sydney NSW 2006, Australia. Electronic address:

Facial landmark detection is to localize multiple facial key-points for a given facial image. While many methods have achieved remarkable performance in recent years, the accuracy remains unsatisfactory due to some uncontrolled conditions such as occlusion, head pose variations and illumination, under which, the L2 loss function is conventionally dominated by errors from those facial components on which the landmarks are hard predicted. In this paper, a novel branched convolutional neural network incorporated with Jacobian deep regression framework, hereafter referred to as BCNN-JDR, is proposed to solve the facial landmark detection problem. Our proposed framework consists of two parts: initialization stage and cascaded refinement stages. We firstly exploit branched convolutional neural networks as the robust initializer to estimate initial shape, which is incorporated with the knowledge of component-aware branches. By virtue of the component-aware branches mechanism, BCNN can effectively alleviate this issue of the imbalance errors among facial components and provide the robust initial face shape. Following the BCNN, a sequence of refinement stages are cascaded to fine-tune the initial shape within a narrow range. In each refinement stage, the local texture information is adopted to fit the facial local nonlinear variation. Moreover, our entire framework is jointly optimized via the Jacobian deep regression optimization strategy in an end-to-end manner. Jacobian deep regression optimization strategy has an ability to backward propagate the training error of the last stage to all previous stages, which implements a global optimization approach to our proposed framework. Experimental results on benchmark datasets demonstrate that the proposed BCNN-JDR is robust against uncontrolled conditions and outperforms the state-of-the-art approaches.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neunet.2019.04.002DOI Listing
October 2019

Parametric Classification of Bingham Distributions Based on Grassmann Manifolds.

IEEE Trans Image Process 2019 Dec 24;28(12):5771-5784. Epub 2019 Jun 24.

In this paper, we present a novel Bayesian classification framework of the matrix variate Bingham distributions with the inclusion of its normalizing constant and develop a consistent general parametric modeling framework based on the Grassmann manifolds. To calculate the normalizing constants of the Bingham model, this paper extends the method of saddle-point approximation (SPA) to a new setting. Furthermore, it employs the standard theory of maximum likelihood estimation (MLE) to evaluate the involved parameters in the used probability density functions. The validity and performance of the proposed approach are tested on 14 real-world visual classification databases. We have compared the classification performance of our proposed approach with the baselines from the previous related approaches. The comparison shows that on most of the databases, the performance of our approach is superior.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2019.2922100DOI Listing
December 2019

Probabilistic Linear Discriminant Analysis With Vectorial Representation for Tensor Data.

IEEE Trans Neural Netw Learn Syst 2019 Oct 21;30(10):2938-2950. Epub 2019 Mar 21.

Linear discriminant analysis (LDA) has been a widely used supervised feature extraction and dimension reduction method in pattern recognition and data analysis. However, facing high-order tensor data, the traditional LDA-based methods take two strategies. One is vectorizing original data as the first step. The process of vectorization will destroy the structure of high-order data and result in high dimensionality issue. Another is tensor LDA-based algorithms that extract features from each mode of high-order data and the obtained representations are also high-order tensor. This paper proposes a new probabilistic LDA (PLDA) model for tensorial data, namely, tensor PLDA. In this model, each tensorial data are decomposed into three parts: the shared subspace component, the individual subspace component, and the noise part. Furthermore, the first two parts are modeled by a linear combination of latent tensor bases, and the noise component is assumed to follow a multivariate Gaussian distribution. Model learning is conducted through a Bayesian inference process. To further reduce the total number of model parameters, the tensor bases are assumed to have tensor CandeComp/PARAFAC (CP) decomposition. Two types of experiments, data reconstruction and classification, are conducted to evaluate the performance of the proposed model with the convincing result, which is superior or comparable against the existing LDA-based methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2019.2901309DOI Listing
October 2019

Fuel-Free Micro-/Nanomotors as Intelligent Therapeutic Agents.

Chem Asian J 2019 Jul 3;14(14):2325-2335. Epub 2019 Apr 3.

School of Materials Science and Engineering, Sun Yat-sen University, Guangzhou, 510275, China.

There are many efficient biological motors in Nature that perform complex functions by converting chemical energy into mechanical motion. Inspired by this, the development of their synthetic counterparts has aroused tremendous research interest in the past decade. Among these man-made motor systems, the fuel-free (or light, magnet, ultrasound, or electric field driven) motors are advantageous in terms of controllability, lifespan, and biocompatibility concerning bioapplications, when compared with their chemically powered counterparts. Therefore, this review will highlight the latest biomedical applications in the versatile field of externally propelled micro-/nanomotors, as well as elucidating their driving mechanisms. A perspective into the future of the micro-/nanomotors field and a discussion of the challenges we need to face along the road towards practical clinical translation of external-field-propelled micro-/nanomotors will be provided.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/asia.201900129DOI Listing
July 2019

Monolithic Structure-Optical Fiber Sensor with Temperature Compensation for Pressure Measurement.

Materials (Basel) 2019 Feb 13;12(4). Epub 2019 Feb 13.

School of Electronic and Information Engineering, Guangdong Ocean University, Zhanjiang 524088, China.

In this paper, an optical fiber pressure sensor cascading a diaphragm-assisted Fabry-Perot interferometer (FPI) and a fiber Bragg grating (FBG) is proposed and demonstrated. The sensor comprises an optical fiber, a fused-silica ferrule, and a fused-silica diaphragm. We use a femtosecond laser firstly to fabricate a pit on the end face of the ferrule and then investigate the laser heat conduction welding and deep penetration welding technology for manufacturing the seepage pressure sensor of the all-fused-silica material. We develop a sensor based on a monolithic structured FPI without adhesive bonding by means of all-laser-welding. The pressure characteristics of the sensor have good linearity at different temperatures. Also, the monolithic structured sensor possesses excellent resolution, hysteresis, and long-term stability. The environmental temperature obtained by the FBG is employed to compensate for the difference in seepage pressure at different temperatures, and the difference in seepage pressure responses at different temperatures is shown to be very small after temperature compensation.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/ma12040552DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6416734PMC
February 2019

Scalable Proximal Jacobian Iteration Method With Global Convergence Analysis for Nonconvex Unconstrained Composite Optimizations.

IEEE Trans Neural Netw Learn Syst 2019 Sep 15;30(9):2825-2839. Epub 2019 Jan 15.

The recent studies have found that the nonconvex relaxation functions usually perform better than the convex counterparts in the l -norm and rank function minimization problems. However, due to the absence of convexity in these nonconvex problems, developing efficient algorithms with convergence guarantee becomes very challenging. Inspired by the basic ideas of both the Jacobian alternating direction method of multipliers (JADMMs) for solving linearly constrained problems with separable objectives and the proximal gradient methods (PGMs) for optimizing the unconstrained problems with one variable, this paper focuses on extending the PGMs to the proximal Jacobian iteration methods (PJIMs) for handling with a family of nonconvex composite optimization problems with two splitting variables. To reduce the total computational complexity by decreasing the number of iterations, we devise the accelerated version of PJIMs through the well-known Nesterov's acceleration strategy and further extend both to solve the multivariable cases. Most importantly, we provide a rigorous convergence analysis, in theory, to show that the generated variable sequence globally converges to a critical point by exploiting the Kurdyka-Łojasiewica (KŁ) property for a broad class of functions. Furthermore, we also establish the linear and sublinear convergence rates of the obtained variable sequence in the objective function. As the specific application to the nonconvex sparse and low-rank recovery problems, several numerical experiments can verify that the newly proposed algorithms not only keep fast convergence speed but also have high precision.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2018.2885699DOI Listing
September 2019

Multiview Subspace Clustering via Tensorial t-Product Representation.

IEEE Trans Neural Netw Learn Syst 2019 Mar 27;30(3):851-864. Epub 2018 Jul 27.

The ubiquitous information from multiple-view data, as well as the complementary information among different views, is usually beneficial for various tasks, for example, clustering, classification, denoising, and so on. Multiview subspace clustering is based on the fact that multiview data are generated from a latent subspace. To recover the underlying subspace structure, a successful approach adopted recently has been sparse and/or low-rank subspace clustering. Despite the fact that existing subspace clustering approaches may numerically handle multiview data, by exploring all possible pairwise correlation within views, high-order statistics that can only be captured by simultaneously utilizing all views are often overlooked. As a consequence, the clustering performance of the multiview data is compromised. To address this issue, in this paper, a novel multiview clustering method is proposed by using t-product in the third-order tensor space. First, we propose a novel tensor construction method to organize multiview tensorial data, to which the tensor-tensor product can be applied. Second, based on the circular convolution operation, multiview data can be effectively represented by a t-linear combination with sparse and low-rank penalty using "self-expressiveness." Our extensive experimental results on face, object, digital image, and text data demonstrate that the proposed method outperforms the state-of-the-art methods for a range of criteria.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2018.2851444DOI Listing
March 2019

Solving Partial Least Squares Regression via Manifold Optimization Approaches.

IEEE Trans Neural Netw Learn Syst 2018 Jul 9. Epub 2018 Jul 9.

Partial least squares regression (PLSR) has been a popular technique to explore the linear relationship between two data sets. However, all existing approaches often optimize a PLSR model in Euclidean space and take a successive strategy to calculate all the factors one by one for keeping the mutually orthogonal PLSR factors. Thus, a suboptimal solution is often generated. To overcome the shortcoming, this paper takes statistically inspired modification of PLSR (SIMPLSR) as a representative of PLSR, proposes a novel approach to transform SIMPLSR into optimization problems on Riemannian manifolds, and develops corresponding optimization algorithms. These algorithms can calculate all the PLSR factors simultaneously to avoid any suboptimal solutions. Moreover, we propose sparse SIMPLSR on Riemannian manifolds, which is simple and intuitive. A number of experiments on classification problems have demonstrated that the proposed models and algorithms can get lower classification error rates compared with other linear regression methods in Euclidean space. We have made the experimental code public at https://github.com/Haoran2014.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2018.2844866DOI Listing
July 2018

Multiview Spectral Clustering via Structured Low-Rank Matrix Factorization.

IEEE Trans Neural Netw Learn Syst 2018 10 4;29(10):4833-4843. Epub 2018 Jan 4.

Multiview data clustering attracts more attention than their single-view counterparts due to the fact that leveraging multiple independent and complementary information from multiview feature spaces outperforms the single one. Multiview spectral clustering aims at yielding the data partition agreement over their local manifold structures by seeking eigenvalue-eigenvector decompositions. Among all the methods, low-rank representation (LRR) is effective, by exploring the multiview consensus structures beyond the low rankness to boost the clustering performance. However, as we observed, such classical paradigm still suffers from the following stand-out limitations for multiview spectral clustering of overlooking the flexible local manifold structure, caused by aggressively enforcing the low-rank data correlation agreement among all views, and such a strategy, therefore, cannot achieve the satisfied between-views agreement; worse still, LRR is not intuitively flexible to capture the latent data clustering structures. In this paper, first, we present the structured LRR by factorizing into the latent low-dimensional data-cluster representations, which characterize the data clustering structure for each view. Upon such representation, second, the Laplacian regularizer is imposed to be capable of preserving the flexible local manifold structure for each view. Third, we present an iterative multiview agreement strategy by minimizing the divergence objective among all factorized latent data-cluster representations during each iteration of optimization process, where such latent representation from each view serves to regulate those from other views, and such an intuitive process iteratively coordinates all views to be agreeable. Fourth, we remark that such data-cluster representation can flexibly encode the data clustering structure from any view with an adaptive input cluster number. To this end, finally, a novel nonconvex objective function is proposed via the efficient alternating minimization strategy. The complexity analysis is also presented. The extensive experiments conducted against the real-world multiview data sets demonstrate the superiority over the state of the arts.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2017.2777489DOI Listing
October 2018

Deep Attention-Based Spatially Recursive Networks for Fine-Grained Visual Recognition.

IEEE Trans Cybern 2019 May 22;49(5):1791-1802. Epub 2018 Mar 22.

Fine-grained visual recognition is an important problem in pattern recognition applications. However, it is a challenging task due to the subtle interclass difference and large intraclass variation. Recent visual attention models are able to automatically locate critical object parts and represent them against appearance variations. However, without consideration of spatial dependencies in discriminative feature learning, these methods are underperformed in classifying fine-grained objects. In this paper, we present a deep attention-based spatially recursive model that can learn to attend to critical object parts and encode them into spatially expressive representations. Our network is technically premised on bilinear pooling, enabling local pairwise feature interactions between outputs from two different convolutional neural networks (CNNs) that correspond to distinct region detection and relevant feature extraction. Then, spatial long-short term memory (LSTMs) units are introduced to generate spatially meaningful hidden representations via the long-range dependency on all features in two dimensions. The attention model is leveraged between bilinear outcomes and spatial LSTMs for dynamic selection on varied inputs. Our model, which is composed of two-stream CNN layers, bilinear pooling, and spatial recursive encoding with attention, is end-to-end trainable to serve as the part detector and feature extractor whereby relevant features are localized, extracted, and encoded spatially for recognition purpose. We demonstrate the superiority of our method over two typical fine-grained recognition tasks: fine-grained image classification and person re-identification.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TCYB.2018.2813971DOI Listing
May 2019

Vectorial Dimension Reduction for Tensors Based on Bayesian Inference.

IEEE Trans Neural Netw Learn Syst 2018 10 21;29(10):4579-4592. Epub 2017 Nov 21.

Dimension reduction for high-order tensors is a challenging problem. In conventional approaches, dimension reduction for higher order tensors is implemented via Tucker decomposition to obtain lower dimensional tensors. This paper introduces a probabilistic vectorial dimension reduction model for tensorial data. The model represents a tensor by using a linear combination of the same order basis tensors, thus it offers a learning approach to directly reduce a tensor to a vector. Under this expression, the projection base of the model is based on the tensor CandeComp/PARAFAC (CP) decomposition and the number of free parameters in the model only grows linearly with the number of modes rather than exponentially. A Bayesian inference has been established via the variational Expectation Maximization (EM) approach. A criterion to set the parameters (a factor number of CP decomposition and the number of extracted features) is empirically given. The model outperforms several existing principal component analysis-based methods and CP decomposition on several publicly available databases in terms of classification and clustering accuracy.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2017.2739131DOI Listing
October 2018

Subspace Clustering via Learning an Adaptive Low-Rank Graph.

IEEE Trans Image Process 2018 Aug;27(8):3716-3728

By using a sparse representation or low-rank representation of data, the graph-based subspace clustering has recently attracted considerable attention in computer vision, given its capability and efficiency in clustering data. However, the graph weights built using the representation coefficients are not the exact ones as the traditional definition is in a deterministic way. The two steps of representation and clustering are conducted in an independent manner, thus an overall optimal result cannot be guaranteed. Furthermore, it is unclear how the clustering performance will be affected by using this graph. For example, the graph parameters, i.e., the weights on edges, have to be artificially pre-specified while it is very difficult to choose the optimum. To this end, in this paper, a novel subspace clustering via learning an adaptive low-rank graph affinity matrix is proposed, where the affinity matrix and the representation coefficients are learned in a unified framework. As such, the pre-computed graph regularizer is effectively obviated and better performance can be achieved. Experimental results on several famous databases demonstrate that the proposed method performs better against the state-of-the-art approaches, in clustering.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2018.2825647DOI Listing
August 2018

Progressive Hard-Mining Network for Monocular Depth Estimation.

IEEE Trans Image Process 2018 Aug;27(8):3691-3702

Depth estimation from the monocular RGB image is a challenging task for computer vision due to no reliable cues as the prior knowledge. Most existing monocular depth estimation works including various geometric or network learning methods lack of an effective mechanism to preserve the cross-border details of depth maps, which yet is very important for the performance promotion. In this paper, we propose a novel end-to-end progressive hard-mining network (PHN) framework to address this problem. Specifically, we construct the hard-mining objective function, the intra-scale and inter-scale refinement subnetworks to accurately localize and refine those hard-mining regions. The intra-scale refining block recursively recovers details of depth maps from different semantic features in the same receptive field while the inter-scale block favors a complementary interaction among multi-scale depth cues of different receptive fields. For further reducing the uncertainty of the network, we design a difficulty-ware refinement loss function to guide the depth learning process, which can adaptively focus on mining these hard-regions where accumulated errors easily occur. All three modules collaborate together to progressively reduce the error propagation in the depth learning process, and then, boost the performance of monocular depth estimation to some extent. We conduct comprehensive evaluations on several public benchmark data sets (including NYU Depth V2, KITTI, and Make3D). The experiment results well demonstrate the superiority of our proposed PHN framework over other state of the arts for monocular depth estimation task.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2018.2821979DOI Listing
August 2018

Log-Euclidean Metrics for Contrast Preserving Decolorization.

IEEE Trans Image Process 2017 Dec 25;26(12):5772-5783. Epub 2017 Aug 25.

This paper presents a novel Log-Euclidean metric inspired color-to-gray conversion model for faithfully preserving the contrast details of color image, which differs from the traditional Euclidean metric approaches. In the proposed model, motivated by the fact that Log-Euclidean metric has promising invariance properties such as inversion invariant and similarity invariant, we present a Log-Euclidean metric-based maximum function to model the decolorization procedure. The Gaussian-like penalty function consisting of the Log-Euclidean metric between gradients of the input color image and transformed grayscale image is incorporated to better reflect the degree of preserving feature discriminability and color ordering in color-to-gray conversion. A discrete searching algorithm is employed to solve the proposed model with linear parametric and non-negative constraints. Extensive evaluation experiments show that the proposed method outperforms the state-of-the-art methods both quantitatively and qualitatively.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2017.2745104DOI Listing
December 2017

On Selecting Effective Patterns for Fast Support Vector Regression Training.

IEEE Trans Neural Netw Learn Syst 2018 08 23;29(8):3610-3622. Epub 2017 Aug 23.

It is time consuming to train support vector regression (SVR) for large-scale problems even with efficient quadratic programming solvers. This issue is particularly serious when tuning the model's parameters. One way to address the issue is to reduce the problem's scale by selecting a subset of the training set. This paper presents a fast pattern selection method by scanning the training data set to reduce a problem's scale. In particular, we find the k-nearest neighbors (kNNs) in a local region around each pattern's target value, and then determine to retain the pattern according to the distribution of its nearest neighbors. There is a high probability that the pattern locates outside the -tube. Since the kNNs of a pattern are found in a very small region, it is fast to scan the whole training data set. The proposed method deals with the year prediction Million Song Data set, which contains 463 715 patterns, within 10 s on a personal computer with an Intel Core i5-4690 CPU at 3.50 GHz and 8GB DRAM. An additional advantage of the proposed method is that it can predefine the size of the selected subset according to the training set. Comprehensive empirical evaluations demonstrate that the proposed method can significantly eliminate redundant patterns for SVR training with only a slight decrease in performance.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2017.2734812DOI Listing
August 2018

Multi-Dimensional Sparse Models.

IEEE Trans Pattern Anal Mach Intell 2018 01 2;40(1):163-178. Epub 2017 Feb 2.

Traditional synthesis/analysis sparse representation models signals in a one dimensional (1D) way, in which a multidimensional (MD) signal is converted into a 1D vector. 1D modeling cannot sufficiently handle MD signals of high dimensionality in limited computational resources and memory usage, as breaking the data structure and inherently ignores the diversity of MD signals (tensors). We utilize the multilinearity of tensors to establish the redundant basis of the space of multi linear maps with the sparsity constraint, and further propose MD synthesis/analysis sparse models to effectively and efficiently represent MD signals in their original form. The dimensional features of MD signals are captured by a series of dictionaries simultaneously and collaboratively. The corresponding dictionary learning algorithms and unified MD signal restoration formulations are proposed. The effectiveness of the proposed models and dictionary learning algorithms is demonstrated through experiments on MD signals denoising, image super-resolution and texture classification. Experiments show that the proposed MD models outperform state-of-the-art 1D models in terms of signal representation quality, computational overhead, and memory storage. Moreover, our proposed MD sparse models generalize the 1D sparse models and are flexible and adaptive to both homogeneous and inhomogeneous properties of MD signals.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2017.2663423DOI Listing
January 2018

Reflectance Prediction Modelling for Residual-Based Hyperspectral Image Coding.

PLoS One 2016 3;11(10):e0161212. Epub 2016 Oct 3.

CM3 Research Unit, School of Computing and Mathematics, Charles Sturt University, Bathurst, NSW, 2795, Australia.

A Hyperspectral (HS) image provides observational powers beyond human vision capability but represents more than 100 times the data compared to a traditional image. To transmit and store the huge volume of an HS image, we argue that a fundamental shift is required from the existing "original pixel intensity"-based coding approaches using traditional image coders (e.g., JPEG2000) to the "residual"-based approaches using a video coder for better compression performance. A modified video coder is required to exploit spatial-spectral redundancy using pixel-level reflectance modelling due to the different characteristics of HS images in their spectral and shape domain of panchromatic imagery compared to traditional videos. In this paper a novel coding framework using Reflectance Prediction Modelling (RPM) in the latest video coding standard High Efficiency Video Coding (HEVC) for HS images is proposed. An HS image presents a wealth of data where every pixel is considered a vector for different spectral bands. By quantitative comparison and analysis of pixel vector distribution along spectral bands, we conclude that modelling can predict the distribution and correlation of the pixel vectors for different bands. To exploit distribution of the known pixel vector, we estimate a predicted current spectral band from the previous bands using Gaussian mixture-based modelling. The predicted band is used as the additional reference band together with the immediate previous band when we apply the HEVC. Every spectral band of an HS image is treated like it is an individual frame of a video. In this paper, we compare the proposed method with mainstream encoders. The experimental results are fully justified by three types of HS dataset with different wavelength ranges. The proposed method outperforms the existing mainstream HS encoders in terms of rate-distortion performance of HS image compression.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0161212PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5047460PMC
June 2017

Tensor LRR and Sparse Coding-Based Subspace Clustering.

IEEE Trans Neural Netw Learn Syst 2016 10 27;27(10):2120-33. Epub 2016 Apr 27.

Subspace clustering groups a set of samples from a union of several linear subspaces into clusters, so that the samples in the same cluster are drawn from the same linear subspace. In the majority of the existing work on subspace clustering, clusters are built based on feature information, while sample correlations in their original spatial structure are simply ignored. Besides, original high-dimensional feature vector contains noisy/redundant information, and the time complexity grows exponentially with the number of dimensions. To address these issues, we propose a tensor low-rank representation (TLRR) and sparse coding-based (TLRRSC) subspace clustering method by simultaneously considering feature information and spatial structures. TLRR seeks the lowest rank representation over original spatial structures along all spatial directions. Sparse coding learns a dictionary along feature spaces, so that each sample can be represented by a few atoms of the learned dictionary. The affinity matrix used for spectral clustering is built from the joint similarities in both spatial and feature spaces. TLRRSC can well capture the global structure and inherent feature information of data and provide a robust subspace segmentation from corrupted data. Experimental results on both synthetic and real-world data sets show that TLRRSC outperforms several established stateof- the-art methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2016.2553155DOI Listing
October 2016

Laplacian Regularized Low-Rank Representation and Its Applications.

IEEE Trans Pattern Anal Mach Intell 2016 Mar;38(3):504-17

Low-rank representation (LRR) has recently attracted a great deal of attention due to its pleasing efficacy in exploring low-dimensional subspace structures embedded in data. For a given set of observed data corrupted with sparse errors, LRR aims at learning a lowest-rank representation of all data jointly. LRR has broad applications in pattern recognition, computer vision and signal processing. In the real world, data often reside on low-dimensional manifolds embedded in a high-dimensional ambient space. However, the LRR method does not take into account the non-linear geometric structures within data, thus the locality and similarity information among data may be missing in the learning process. To improve LRR in this regard, we propose a general Laplacian regularized low-rank representation framework for data representation where a hypergraph Laplacian regularizer can be readily introduced into, i.e., a Non-negative Sparse Hyper-Laplacian regularized LRR model (NSHLRR). By taking advantage of the graph regularizer, our proposed method not only can represent the global low-dimensional structures, but also capture the intrinsic non-linear geometric information in data. The extensive experimental results on image clustering, semi-supervised image classification and dimensionality reduction tasks demonstrate the effectiveness of the proposed method.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2015.2462360DOI Listing
March 2016