Jing-Hua Yang

To deal with the curse of dimensionality in hyperspectral images, numerous feature dimensionality reduction (FDR) methods have been proposed to map high-dimensional data into a low-dimensional subspace. However, most of existing FDR methods lack robustness against noise corruption. To this end, the representation-based subspace learning has been developed to find a robust projection matrix for FDR. Nevertheless, most of them only consider a single direction of the matrix, which ignore the information from other directions. Moreover, the majority of existing methods fail to account for both global structure and feature correlations effectively. To address the above problems, we propose a novel robust projection learning method called latent low-rank embedding (LatLRE), which integrates the latent low-rank representation (LatLRR) with projection learning. In particular, the proposed model can maintain the strong robustness of LatLRR and simultaneously learn a projection for FDR. Moreover, the nuclear norm and logarithmic norm are employed to approximate the two underlying rank functions and provide a more accurate measure of correlation. In addition, LatLRE is optimized using the alternating direction method of multipliers (ADMM) algorithm with the theoretical convergence guarantee. To verify the FDR performance of LatLRE, extensive experiments are conducted on three benchmark hyperspectral datasets. The experimental results demonstrate that LatLRE outperforms other FDR methods considered in this paper.

Remote sensing image (RSI) denoising is an important and fundamental task in RSI processing. Existing denoising methods usually assume that RSI lies in a single matrix or tensor subspace. However, due to the wavelength difference or/and temporal variability, the assumption of a single subspace may not be suitable for RSI. To address this, we propose a tensor multi-subspace representation (TenMSR) for RSI mixed noise removal. To be specific, in this work, we introduce TenMSR to finely characterize the intrinsic tensor multi-subspace structure of RSI. Compared with the single matrix/tensor subspace-based methods, the proposed method can not only precisely describe the wavelength difference or/and temporal variability of RSI but also produce a more compact image distribution in tensor multi-subspace. To mine and preserve the multi-subspace structure, we introduce a nonlinear transform-based 3-D tensor nuclear norm to characterize the tensor low rankness of the multi-subspace representation coefficient. An effective algorithm based on the proximal alternating minimization (PAM) framework is developed to solve the proposed model with theoretical convergence analysis. Extensive experiments show the effectiveness and superiority of the proposed method over existing state-of-the-art single matrix/tensor subspace RSI denoising methods.

In recent years, graph convolutional networks (GCNs) have gained increasing attention in hyperspectral image (HSI) classification due to their good ability to model the pairwise relationships between two pixels. However, it is difficult to effectively model more complex relationships among multiple pixels with simple graphs. To solve this problem, we propose a novel tensorized high-order hypergraph convolutional network (TH²GCN) for HSI classification. Specifically, the hypergraph structure is employed to effectively model complex spatial relationships between pixels in HSIs, and we propose a new tensor-based algebraic representation of hypergraphs as a powerful strategy for describing the high-order interaction structures of the hypergraph. Besides, by extending the adjacency matrix-based GCN to the tensor domain and exploiting the tensor decomposition, the TH²GCN method is designed to efficiently extract high-order discriminative information from the hypergraph at low complexity for improving HSI classification performance. Furthermore, the construction of the adjacency tensor on all the data requires a huge amount of memory, especially for large-scale remote sensing images. To this end, the TH²GCN is trained and tested for HSI data in a minibatch fashion. Experimental results on three HSI datasets prove that the performance of the proposed method outperforms the comparison methods.

By focusing on the structure exploration and information propagation from non-Euclidean data space, graph convolutional neural network (GCN), which can extract abundant and discriminative features, has been a valuable topic in polarimetric synthetic aperture radar (PolSAR) image field. However, the existing GCN-based PolSAR classification methods have high computational cost, may easily be prone to over-smoothing or over-fitting, and inadequately learn the polarimetric property. To address these issues, we propose a polarimetric rotation-based multiscale Gaussian process-driven GCN (MGPGCN) for semi-supervised PolSAR image classification. First, for addressing the over-smoothing and over-fitting problems, the Gaussian process (GP) is introduced into GCN framework, which can fit the underlying feature distribution rather than calculating specific values of weights in conventional GCN. Second, we extend the multiscale layer architecture and design the multiscale kernel for improving the representation capability and fully leveraging neighborhood information of GCN. Third, to mitigate the effect caused by noise or imaging angle, a superpixel-level polarimetric rotation-based feature enhancement strategy is designed. With this strategy, the characteristic of each terrain type is more salient, and the representation capability of GP kernel can be further improved. Comprehensive experiments on four PolSAR datasets firmly demonstrate that the proposed MGPGCN can achieve better performance compared with some widely-used GCN-based classification methods.

Multi-view subspace clustering (MVSC) separates the data with multiple views into multiple clusters, and each cluster corresponds to one certain subspace. Existing tensor-based MVSC methods construct self-representation subspace coefficient matrices of all views as a tensor, and introduce the tensor nuclear norm (TNN) to capture the complementary information hidden in different views. The key assumption is that the data samples of each subspace must be sufficient for subspace representation. This work proposes a nonconvex latent transformed low-rank tensor representation framework for MVSC. To deal with the insufficient sample problem, we study the latent low-rank representation in the multi-view case to supplement underlying observed samples. Moreover, we propose to use data-driven transformed TNN (TTNN), resulting from the intrinsic structure of multi-view samples, to preserve the consensus and complementary information in the transformed domain. Meanwhile, the proposed unified nonconvex low-rank tensor representation framework can better learn the high correlation among different views. To resolve the proposed nonconvex optimization model, we propose an effective algorithm under the framework of the alternating direction method of multipliers and theoretically prove that the iteration sequences converge to the critical point. Experiments on various datasets showcase outstanding performance.

With the development of the theory and technology of computer science, machine or computer painting is increasingly being explored in the creation of art. Machine-made works are referred to as artificial intelligence (AI) artworks. Early methods of AI artwork generation have been classified as non-photorealistic rendering, and, latterly, neural style transfer methods have also been investigated. As technology advances, the variety of machine-generated artworks and the methods used to create them have proliferated. However, there is no unified and comprehensive system to classify and evaluate these works. To date, no work has generalized methods of creating AI artwork including learning-based methods for painting or drawing. Moreover, the taxonomy, evaluation, and development of AI artwork methods face many challenges. This article is motivated by these considerations. We first investigate current learning-based methods for making AI artworks and classify the methods according to art styles. Furthermore, we propose a consistent evaluation system for AI artworks and conduct a user study to evaluate the proposed system on different AI artworks. This evaluation system uses six criteria: beauty, color, texture, content detail, line, and style. The user study demonstrates that the six-dimensional evaluation index is effective for different types of AI artworks.

Most multi-view clustering methods based on shallow models are limited in sound nonlinear information perception capability, or fail to effectively exploit complementary information hidden in different views. To tackle these issues, we propose a novel Subspace-Contrastive Multi-View Clustering (SCMC) approach. Specifically, SCMC utilizes a set of view-specific auto-encoders to map the original multi-view data into compact features capturing its nonlinear structures. Considering the large semantic gap of data from different modalities, we project multiple heterogeneous features into a joint semantic space, namely the embedded compact features are passed through the self-expression layers to learn the subspace representations, respectively. In order to enhance the discriminability and efficiently excavate the complementarity of various subspace representations, we use the contrastive strategy to maximize the similarity between positive pairs while differentiate negative pairs. Thus, the graph regularization is employed to encode the local geometric structure within varying subspaces for optimizing the consistent affinity matrix. Furthermore, to endow the proposed SCMC with the ability of handling the multi-view out-of-samples, we develop a consistent sparse representation (CSR) learning mechanism over the in-samples. To demonstrate the effectiveness of the proposed model, we conduct a large number of comparative experiments on ten challenging datasets, and the experimental results show that SCMC outperforms existing shallow and deep multi-view clustering methods. In addition, the experimental results on out-of-samples illustrate the effectiveness of the proposed CSR.