检索结果-内蒙古大学图书馆

Embanet: A Flexible Efficient Multi-Branch Attention Network

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

This work presents a novel module, namely multi-branch concat (MBC), toprocess the input tensor and obtain the multi-scale feature map. The proposedMBC module brings new degrees of freedom (DoF) for the design of attentionnetworks by allowing the type of transformation operators and the numberof branches to be flexibly adjusted. Two important transformation operators,multiplex and split, are considered in this work, both of which can representmulti-scale features at a more granular level and increase the range of receptivefields. By integrating the MBC and attention module, a multi-branch attention(MBA) module is consequently developed to capture the channel-wise interactionof feature maps for establishing the long-range channel dependency. By substituting the 3x3 convolutions in the bottleneck blocks of the ResNet with theproposed MBA, a novel block namely efficient multi-branch attention (EMBA)is obtained, which can be easily plugged into the state-of-the-art backbone CNNmodels. Furthermore, a new backbone network called EMBANet is establishedby stacking the EMBA blocks. The proposed EMBANet is extensively evaluatedon representative computer vision tasks including: classification, detection, andsegmentation. And it demonstrates consistently superior performance over thepopular backbones. © 2023, The Authors. All rights reserved.

关键词：

Atomic Filter: a Weak Form of Shift Operator for Graph Signals

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yang, Lihua Zhang, Qing Zhang, Qian Huang, Chao School of Mathematics Sun Yat-sen University Guangzhou510275 China Guangdong Province Key Laboratory of Computational Science China College of Mathematics and Statistics Shenzhen University Shenzhen518060 China Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen518060 China

The shift operation plays a crucial role in the classical signal processing. It is the generator of all the filters and the basic operation for time-frequency analysis, such as windowed Fourier transform and wavelet transform. With the rapid development of internet technology and big data science, a large amount of data are expressed as signals defined on graphs. In order to establish the theory of filtering, windowed Fourier transform and wavelet transform in the setting of graph signals, we need to extend the shift operation of classical signals to graph signals. It is a fundamental problem since the vertex set of a graph is usually not a vector space and the addition operation cannot be defined on the vertex set of the graph. In this paper, based on our understanding on the core role of shift operation in classical signal processing we propose the concept of atomic filters, which can be viewed as a weak form of the shift operator for graph signals. Then, we study the conditions such that an atomic filter is norm-preserving, periodic, or real-preserving. The property of real-preserving holds naturally in the classical signal processing, but no the research has been reported on this topic in the graph signal setting. With these conditions we propose the concept of normal atomic filters for graph signals, which degenerates into the classical shift operator under mild conditions if the graph is circulant. Typical examples of graphs that have or have not normal atomic filters are given. Finally, as an application, atomic filters are utilized to construct time-frequency atoms which constitute a frame of the graph signal space. Copyright © 2022, The Authors. All rights reserved.

关键词： Wavelet transforms

Feature Selection for Data Classification based on Binary Brain Storm Optimization

学校读者我要写书评

暂无评论

Feature Selection for Data Classification based on Binary Br...

IEEE International Conference on Cloud Computing and Intelligence Systems (CCIS)

作者： Farhad Pourpanah Ran Wang Xizhao Wang College of Mathematics and Statistics Shenzhen University China Shenzhen Key Lab. Of advanced machine learning and applications Shenzhen University China Guangdong Key Lab. of Intelligent Information Processing Shenzhen University China

ISBN: (数字)9781728138633

ISBN: (纸本)9781728138640

Brain Storm Optimization (BSO) is an effective population-based optimization method inspired by human brainstorming process. This paper proposes a new binary BSO algorithm (BBSO) to develop a new feature selection approach in order to reduce the number of selected features and/or improve the classification accuracy. Specifically, a new update rule mechanism for generating new solutions is proposed, which improves the convergence speed and reduces the possibility of immature convergence in the algorithm. Furthermore, a fuzzy ARTMAP (FAM) neural network, which is an incremental learning model, is utilized as a classification approach to evaluate the effectiveness of the selected feature subsets. The performance of the proposed method is compared with those from the original BSO, particle swarm optimization (PSO) and genetic algorithm (GA) on eight commonly used and well-known benchmark problems. The experimental results indicate the superiority of the BBSO as compared with other state-of-the-art feature selection methods.

关键词： Knowledge based systems

Orthogonal subspace based fast iterative thresholding algorithms for joint sparsity recovery

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Han, Ningning Li, Shidong Lu, Jian Shenzhen Key Laboratory of Advanced Machine Learning and Applications College of Mathematics and Statistics Shenzhen University Shenzhen518060 China Department of Mathematics San Francisco State University San FranciscoCA94132 United States

Sparse signal recoveries from multiple measurement vectors (MMV) with joint sparsity property have many applications in signal, image, and video processing. The problem becomes much more involved when snapshots of the signal matrix are temporally correlated. With signal’s temporal correlation in mind, we provide a framework of iterative MMV algorithms based on thresholding, functional feedback and null space tuning. Convergence analysis for exact recovery is established. Unlike most of iterative greedy algorithms that select indices in a measurement/solution space, we determine indices based on an orthogonal subspace spanned by the iterative sequence. In addition, a functional feedback that controls the amount of energy relocation from the "tails" is implemented and analyzed. It is seen that the principle of functional feedback is capable to lower the number of iteration and speed up the convergence of the algorithm. Numerical experiments demonstrate that the proposed algorithm has a clearly advantageous balance of efficiency, adaptivity and accuracy compared with other state-of-the-art algorithms. © 2021, CC BY.

关键词： Vectors

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Xu Jia, Yuheng Song, Mofei Wang, Ran School of Computer Science and Engineering South University Nanjing210093 China School of Mathematical Sciences Shenzhen University Shenzhen518060 China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen518060 China Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen518060 China

Ensemble clustering aggregates multiple weak clusterings to achieve a more accurate and robust consensus result. The Co-Association matrix (CA matrix) based method is the mainstream ensemble clustering approach that constructs the similarity relationships between sample pairs according the weak clustering partitions to generate the final clustering result. However, the existing methods neglect that the quality of cluster is related to its size, i.e., a cluster with smaller size tends to higher accuracy. Moreover, they also do not consider the valuable dissimilarity information in the base clusterings which can reflect the varying importance of sample pairs that are completely disconnected. To this end, we propose the Similarity and Dissimilarity Guided Co-association matrix (SDGCA) to achieve ensemble clustering. First, we introduce normalized ensemble entropy to estimate the quality of each cluster, and construct a similarity matrix based on this estimation. Then, we employ the random walk to explore high-order proximity of base clusterings to construct a dissimilarity matrix. Finally, the adversarial relationship between the similarity matrix and the dissimilarity matrix is utilized to construct a promoted CA matrix for ensemble clustering. We compared our method with 13 state-of-the-art methods across 12 datasets, and the results demonstrated the superiority clustering ability and robustness of the proposed approach. The code is available at https://***/xuz2019/SDGCA. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

ORTHOGONAL CONSTRAINED MINIMIZATION WITH TENSOR 2,p REGULARIZATION FOR HSI DENOISING AND DESTRIPING

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Xiaoxia Yu, Shijie Lu, Jian Chen, Xiaojun Shenzhen Key Laboratory of Advanced Machine Learning and Applications School of Mathematical Sciences Shenzhen University Shenzhen518060 China Department of Applied Mathematics The Hong Kong Polytechnic University Hong Kong National Center for Applied Mathematics Shenzhen Shenzhen518055 China

Hyperspectral images (HSIs) are often contaminated by a mixture of noises such as Gaussian noise, dead lines, stripes, and so on. In this paper, we propose a novel approach for HSI denoising and destriping, called MLTL2p, which consists of an orthogonal constrained minimization model and an iterative algorithm with convergence guarantees. The model of the proposed MLTL2p approach is built based on a new sparsity-enhanced Multi-scale Low-rank Tensor regularization and a tensor 2,p norm with p ∈ (0, 1). The multi-scale low-rank regularization for HSI denoising utilizes the global and local spectral correlation as well as the spatial nonlocal self-similarity priors of HSIs. The corresponding low-rank constraints are formulated based on independent higher-order singular value decomposition with sparsity enhancement on its core tensor to prompt more low-rankness. The tensor 2,p norm for HSI destriping is extended from the matrix 2,p norm. A proximal block coordinate descent algorithm is proposed in the MLTL2p approach to solve the resulting nonconvex nonsmooth minimization with orthogonal constraints. We show any accumulation point of the sequence generated by the proposed algorithm converges to a first-order stationary point, which is defined using three equalities of substationarity, symmetry, and feasibility for orthogonal constraints. In the numerical experiments, we compare the proposed method with state-of-the-art methods including a deep learning based method, and test the methods on both simulated and real HSI datasets. Our proposed MLTL2p method demonstrates outperformance in terms of metrics such as mean peak signal-to-noise ratio as well as visual *** Codes 68U10, 90C26, 15A18, 65F22 Copyright © 2024, The Authors. All rights reserved.

关键词： Signal to noise ratio

Geometric Edge Convolution for Rigid Transformation Invariant Features in 3d Point Clouds

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Bello, Saifullahi Aminu Alfasly, Saghir Mao, Jiawei Lu, Jian Li, Lin Xu, Chen Zou, Yuru Shenzhen Key Laboratory of Advanced Machine Learning and Applications College of Mathematics and Statistics Shenzhen University Shenzhen518060 China Shenzhen518055 China Pazhou Lab Guangzhou510335 China School of Electronic Engineering Xidian University Xi’an710071 China

Extracting rigid transformation invariant features is still a challenge on 3D point clouds because rigid transformation changes the point coordinates, and relying on the point coordinates, most existing deep learning models fail. This paper addresses this challenge by proposing geometric edge convolution (GEConv). To facilitate rigid transformation invariant feature extraction, GEConv uses geometric features to construct a local graph and use shared multi-layer perceptron (MLP) on the graph edges to extract deep features. By stacking GEConv, we construct a deep model called GEConvNet that hierarchically extracts different level rigid transformation invariant features for various point cloud tasks. To prove the effectiveness of GEConv, we provide experiments on four point cloud processing tasks;classification, parts segmentation, semantic segmentation, and point set registration. In the classification and segmentation tasks, GEConvNet performs better on robustness to rotation compared to existing models. Furthermore, our feature-matching-based approach for point set registration demonstrates strong performance, providing further evidence of GEConv's ability to handle rotation and translation effectively. Our code will be made available on http://***/saifabel/GEConv © 2024, The Authors. All rights reserved.

关键词： Feature extraction

HOCA: Higher-Order Channel Attention for Single Image Super-Resolution

学校读者我要写书评

暂无评论

HOCA: Higher-Order Channel Attention for Single Image Super-...

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Yalei Lv Tao Dai Bin Chen Jian Lu Shu-Tao Xia Jingchao Cao Tsinghua Shenzhen International Graduate School Tsinghua University Shenzhen China PCL Research Center of Networks and Communications Peng Cheng Laboratory Shenzhen China Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University China City University of Hong Kong Hong Kong

Convolutional neural networks (CNNs) have obtained great success in single image super-resolution (SR). More recent works (e.g., RCAN and SAN) have obtained remarkable performance with channel attention based on first- or second-order statistics of features. However, these methods neglect the rich feature statistics higher than second-order, thus hindering the representation ability of CNNs. To address this issue, we propose a higher-order channel attention (HOCA) module to enhance the representation ability of CNNs. In our HOCA module, to capture different types of semantic information, we first compute k-order of feature statistics, followed by channel attention to learn the feature interdependencies. Considering the diversity of input contents, we design a gate mechanism to adaptively select a specific k-order channel attention. Besides, our HOCA module serves as a plug-and-play module and can be easily plugged into existing state-of-art CNN-based SR methods. Extensive experiments on public benchmarks show that our HOCA module effectively improves the performance of various CNN-based SR methods.

关键词： Adaptation models Conferences Superresolution Semantics Signal processing Logic gates Benchmark testing

RIDNet: Recursive Information Distillation Network for Color Image Denoising

学校读者我要写书评

暂无评论

RIDNet: Recursive Information Distillation Network for Color...

International Conference on Computer Vision Workshops (ICCV Workshops)

作者： Shengkai Zhuo Zhi Jin Wenbin Zou Xia Li Shenzhen University China College of Electronics and Information Engineering Shenzhen Key Laboratory of Advanced Machine Learning and Applications Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University School of Intelligent Systems Engineering Sun Yat-sen University Sun Yat-sen University China

ISBN: (数字)9781728150239

ISBN: (纸本)9781728150246

Color image denoising is more challenging in effectiveness when compared with the grayscale one. Most existing methods play a certain role in efficiency or flexibility, but lack robustness to handle various noise levels, especially the severe noise. This keeps them away from being practically applied to color image denoising. To address this issue, we propose a robust CNN based denoiser, namely Recursive Information Distillation Network (RIDNet), to handle the denoising task at high noise levels. The proposed RIDNet simultaneously keeps the efficiency and flexibility by introducing the information distillation module and merging a tunable noise level map as the input, respectively. Experiment results on Additive White Gaussian Noise (AWGN) images demonstrate that our method outperforms most of the state-of-the-art color image denoisers.

关键词： Noise reduction Noise level Color Training AWGN Noise measurement Task analysis