21 June 2022 in IEEE Transactions on Knowledge and Data Engineering
Multi-view clustering method tries to improve the performance of clustering by using the information existing in different views. The tensorial representation is more suitable to capture the high order correlations across different views while keep local geometrical structure in specific view. In this paper, we propose a sliced sparse gradient induced multi-view subspace clustering method via tensorial arctangent rank minimization, named SSG-TAR method. Firstly, a tensorial arctangent rank (TAR) is defined, which is a tighter surrogate of the tensor rank and more effective to explore the consistency among multiple views. Secondly, a sliced sparse gradient regularization (SSG) is firstly proposed to enhance the discrimination between clusters and better capture the complementary information in view-specific feature space. Finally, we unify these two terms together and establish an efficient algorithm to optimize the proposed model. Furthermore, the constructed sequence was proved to converge to the stationary KKT point. We have carried out extensive experiments on ten datasets across different types and sizes to verify the performance of our model. The experimental results show that our method have achieved the state-of-the-art performance.
30 August 2021 in IEEE Transactions on Neural Networks and Learning Systems
Saliency detection is an important but challenging task in the study of computer vision. In this article, we develop a new unsupervised learning approach for the saliency detection by an intrinsic regularization model, in which the Schatten-2/3 norm is integrated with the nonconvex sparse ${l_{2/3}}$ norm. The ${l_{2/3}}$ -norm is shown to be capable of detecting consistent values among sparse foreground by using image geometrical structure and feature similarity, while the Schatten-2/3 norm can capture the lower rank of background by matrix factorization. To improve effective performance of separation for Schatten-2/3-norm and ${l_{2/3}}$ -norm, a Laplacian regularization is adopted to the foreground for the smoothness. The proposed model essentially converts the required nonconvex optimization problem into the convex one, conducted by splitting the objective function based on singular value decomposition on one much smaller factor matrix and then optimized by using the alternating direction method of the multiplier. The convergence of the proposed algorithm is discussed in detail. Extensive experiments on three benchmark datasets demonstrate that our unsupervised learning approach is very competitive and appears to be more consistent across various salient objects than the current existing approaches.
13 January 2022 in IEEE Transactions on Cybernetics
In the study of salient object detection, multiview features play an important role in identifying various underlying salient objects. As to current common patch-based methods, all different features are handled directly by stacking them into a high-dimensional vector to represent related image patches. These approaches ignore the correlations inhering in the original spatial structure, which may lead to the loss of certain underlying characterization such as view interaction. In this article, different from currently available approaches, a tensorial feature representation framework is developed for the salient object detection in order to better explore the complementary information of multiview features. Under the tensor framework, a tensor low-rank constraint is applied to the background to capture its intrinsic structure, a tensor group sparsity regularization is posed on the salient part, and a tensorial sliced Laplacian regularization is then introduced to enlarge the gap between the subspaces of the background and salient object. Moreover, a nonconvex tensor Log-determinant function, instead of the tensor nuclear norm, is adopted to approximate the tensor rank for effectively suppressing the confusing information resulted from underlying complex backgrounds. Further, we have deduced the closed-form solution of this nonconvex minimization problem and established a feasible algorithm whose convergence is mathematically proven. Experiments on five well-known public datasets are provided and the simulations demonstrate that our method outperforms the latest unsupervised handcrafted features-based methods in the literature. Furthermore, our model is flexible with various deep features and is competitive with the state-of-the-art approaches.
1 July 2023 in Chinese Journal of Electronics
Multi-view subspace clustering under a tensor framework remains a challenging problem, which can be potentially applied to image classification, impainting, denoising, etc. There are some existing tensor-based multi-view subspace clustering models mainly making use of the consistency in different views through tensor nuclear norm (TNN). The diversity which means the intrinsic difference in individual view is always ignored. In this paper, a new tensorial multi-view subspace clustering model is proposed, which jointly exploits both the consistency and diversity in each view. The view representation is decomposed into view-consistent part (low-rank part) and view-specific part (diverse part). A tensor adaptive log-determinant regularization (TALR) is imposed on the low-rank part to better relax the tensor multi-rank, and a view-specific sparsity regularization is applied on the diverse part to ensure connectedness property. Although the TALR minimization is not convex, it has a closed-form analytical solution and its convergency is validated mathematically. Extensive evaluations on six widely used clustering datasets are executed and our model is demonstrated to have the superior performance.
2 June 2021 in Connection Science
Recently, blur detection is a hot topic in computer vision. It can accurately segment the blurred areas from an image, which is conducive for the post-processing of the image. Although many hand-crafted features based approaches have been presented during the last decades, they were not robust to the complex scenarios. To solve this problem, we newly establish a boundary-aware multi-scale deep network in this paper. First, the VGG-16 network is used to extract the deep features from multi-scale layers. Contrast layers and deconvolutional layers are added to make the difference between the blurred areas and clear areas more prominent. At last, a new boundary-aware penalty is introduced, which makes the edges of our results much clearer. Our method spends about 0.2 s to evaluate an image. Experiments on the large dataset confirm that the proposed model performs better than other models.
19 August 2020 in Semigroup Forum