检索结果-内蒙古大学图书馆

Successive Consensus Clustering for Unsupervised Video-Based Person Re-Identification

IEEE signal processing LETTERS 2022年 29卷 822-826页

作者： Qian, Jinhao Xie, Xiaohua Sun Yat Sen Univ Guangdong Prov Key Lab Informat Secur Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China Sun Yat Sen Univ Minist Educ Key Lab Machine Intelligence & Adv Comp Guangzhou 510006 Peoples R China

Person re-identification is to match the same person between non-overlapping cameras. This paper focuses on unsupervised video-based person re-identification. The mainstream approach is to obtain pseudo-labels by clustering samples for training the classification model. In this scheme, a potential threat is that noisy pseudo-labels may damage the optimization of the model. To mitigate this danger, we propose using a Successive Consensus Clustering framework for optimizing the pseudo-labels and the model iteratively. First, we leverage consensus clustering with respect to multiple frames of a video, which can generate high-quality pseudo-labels for pedestrians. Secondly, we develop contrastive learning based on the cluster successive memory mechanism, which can establish the correlation between different epochs of clustering so that makes the training of the model stable. Experiments on three large-scale data sets show that our method outperforms the previous state-of-the-art method, surpassing 10.6% for rank-1 and 18.6% for mAP on Mars, and 9.6% for rank-1 and 13.3% for mAP on DukeMTMC-VideoReID.

关键词： Feature extraction Training Clustering algorithms Optimization signal processing algorithms Neural networks Memory modules Unsupervised person re-identification consensus clustering contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Efficient Tensor Robust PCA Under Hybrid Model of Tucker and Tensor Train

引用

IEEE signal processing LETTERS 2022年 29卷 627-631页

作者： Qiu, Yuning Zhou, Guoxu Huang, Zhenhao Zhao, Qibin Xie, Shengli Guangdong Univ Technol Sch Automat Guangzhou 510006 Peoples R China Guangdong Univ Technol Key Lab Intelligent Detect & Internet Things Mfg Guangzhou 510006 Peoples R China Guangdong Univ Technol Minist Educ Guangzhou 510006 Peoples R China RIKEN Ctr Adv Intelligence Project AIP Tokyo 1030027 Japan Guangdong Univ Technol Guangdong Hong Kong Macao Joint Lab Smart Discret Guangzhou 510006 Peoples R China

Tensor robust principal component analysis (TRPCA) is a fundamental model in machine learning and computer vision. Recently, tensor train (TT) decomposition has been verified effective to capture the global low-rank correlation for tensor recovery tasks. However, due to the large-scale tensor data in real-world applications, existing TRPCA models often suffer from high computational complexity. In this letter, we propose an efficient TRPCA under hybrid model of Tucker and TT. Specifically, in theory we reveal that TT nuclear norm (TTNN) of the original big tensor can be equivalently converted to that of a much smaller tensor via a Tucker compression format, thereby significantly reducing the computational cost of singular value decomposition (SVD). Numerical experiments on both synthetic and real-world tensor data verify the superiority of the proposed model.

关键词： Tensors Matrix decomposition Computational modeling Optimization Minimization Matrix converters signal processing algorithms Tensor analysis robust tensor decomposition tensor train decomposition tensor robust principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Gaussian Process Convolutional Dictionary Learning

引用

IEEE signal processing LETTERS 2022年 29卷 95-99页

作者： Song, Andrew H. Tolooshams, Bahareh Ba, Demba MIT Dept Elect Engn & Comp Sci Cambridge MA 02139 USA Harvard Univ Sch Engn & Appl Sci Allston MA 02134 USA

Convolutional dictionary learning (CDL), the problem of estimating shift-invariant templates from data, is typically conducted in the absence of a prior/structure on the templates. In data-scarce or low signal-to-noise ratio (SNR) regimes, learned templates overfit the data and lack smoothness, which can affect the predictive performance of downstream tasks. To address this limitation, we propose GPCDL, a convolutional dictionary learning framework that enforces priors on templates using Gaussian Processes (GPs). With the focus on smoothness, we show theoretically that imposing a GP prior is equivalent to Wiener filtering the learned templates, thereby suppressing high-frequency components and promoting smoothness. We show that the algorithm is a simple extension of the classical iteratively reweighted least squares algorithm, independent of the choice of GP kernels. This property allows one to experiment flexibly with different smoothness assumptions. Through simulation, we show that GPCDL learns smooth dictionaries with better accuracy than the unregularized alternative across a range of SNRs. Through an application to neural spiking data, we show that GPCDL learns a more accurate and visually-interpretable smooth dictionary, leading to superior predictive performance compared to non-regularized CDL, as well as parametric alternatives.

关键词： Dictionaries Kernel Convolution signal to noise ratio Machine learning Gaussian processes signal processing algorithms Convolutional dictionary learning Gaussian process exponential family Wiener filter smoothness

来源：评论

学校读者我要写书评

暂无评论

Covariance-Based Joint Device Activity and Delay Detection in Asynchronous mMTC

引用

IEEE signal processing LETTERS 2022年 29卷 538-542页

作者： Wang, Zhaorui Liu, Ya-Feng Liu, Liang Chinese Univ Hong Kong Dept Informat Engn Hong Kong Peoples R China Chinese Acad Sci Acad Math & Syst Sci Inst Computat Math & Sci Engn Comp State Key Lab Sci & Engn Comp Beijing 100190 Peoples R China Hong Kong Polytech Univ Dept Elect & Informat Engn Hong Kong Peoples R China

In this letter, we study the joint device activity and delay detection problem in asynchronous massive machine-type communications (mMTC), where all active devices asynchronously transmit their preassigned preamble sequences to the base station (BS) for device identification and delay detection. We first formulate this joint detection problem as a maximum likelihood estimation problem, which depends on the received signal only through its sample covariance, and then propose efficient coordinate descent type of algorithms to solve the formulated problem. Our proposed covariance-based approach is sharply different from the existing compressed sensing (CS) approach for the same problem. Numerical results show that our proposed covariance-based approach significantly outperforms the CS approach in terms of the detection performance since our proposed approach can make better use of the BS antennas than the CS approach.

关键词： Delays Channel estimation signal processing algorithms Synchronization Protocols Performance evaluation Optimization Asynchronous mMTC coordinate descent joint activity and delay detection random access

来源：评论

学校读者我要写书评

暂无评论

Novel Criterion for Preventing Overflow Oscillations in Fixed-Point Digital Filters With State Saturation

引用

IEEE signal processing LETTERS 2022年 29卷 1287-1291页

作者： Agarwal, Neha Kar, Haranath Motilal Nehru Natl Inst Technol Elect & Commun Engn Dept Prayagraj 211004 India

This letter presents a novel global asymptotic stability (GAS) criterion for fixed-point digital filters (DFs) employing saturation overflow arithmetic. The criterion is developed using an improved characterization of saturation nonlinearities (as prevailing in the DF under study) and Lyapunov theory. The criterion outperforms a series of existing criteria.

关键词： Asymptotic stability Stability criteria Arithmetic Symmetric matrices signal processing algorithms Digital filters Trajectory Digital filter global asymptotic stability linear matrix inequality Lyapunov method saturation arithmetic

来源：评论

学校读者我要写书评

暂无评论

A Low Complexity Symbol-Wise ML Detection Algorithm for User-Centric C-RAN

引用

IEEE COMMUNICATIONS LETTERS 2022年第5期26卷 1057-1061页

作者： Wang, Xinyi Liu, Wenlong Jin, Minglu Dalian Univ Technol Sch Informat & Commun Engn Dalian 116024 Peoples R China

This letter considers an optimal signal detection problem for uplink user-centric cloud radio access network (C-RAN). In user-centric C-RAN, users communicate with nearby remote radio heads (RRHs), and RRHs are connected to baseband unit (BBU) pool. Then, the baseband processing functionalities are migrated to the BBU pool to provide joint signal detection for a number of users. However, joint signal processing will encounter high computational complexity as the network scale grows. Especially for optimal signal detection, the computational complexity increases exponentially as the number of users increases. To address this issue, we propose a symbol-wise maximum likelihood (ML) detection algorithm to achieve optimal performance with low complexity. The key to realize the symbol-by-symbol separate detection is that it makes full use of the sparsity of the channel matrix in user-centric C-RAN. We further provide the theoretical derivation to clarify the superiority of the proposed algorithm in this sparse channel scenario. Numerical experiments show that the proposed algorithm can achieve optimal performance with low complexity in user-centric C-RAN.

关键词： Detection algorithms Sparse matrices signal processing algorithms signal detection Detectors Iterative algorithms Computational complexity C-RAN user-centric signal detection ML detector optimal performance

来源：评论

学校读者我要写书评

暂无评论

NLMS is More Robust to Input-Correlation Than LMS: A Proof

引用

IEEE signal processing LETTERS 2022年 29卷 279-283页

作者： Ali, Anum Moinuddin, Muhammad Al-Naffouri, Tareq Y. Samsung Res Amer Stand & Mobil Innovat Lab Piano TX 75023 USA King Abdulaziz Univ Dept Elect & Comp Engn Jeddah 21589 Saudi Arabia King Abdulaziz Univ Ctr Excellence Intelligent Engn Syst CEIES Jeddah 21589 Saudi Arabia King Abdullah Univ Sci & Technol Comp Elect & Math Sci & Engn Div Thuwal 239556900 Saudi Arabia

In this work, we comparatively analyze the least mean squares (LMS) algorithm and the normalized least mean squares (NLMS) algorithm. We use the input moment matrices for comparison as the mean-square behavior of both algorithms is determined by the input moment matrices. First, we derive the closed-form expressions of the input moment matrices of the NLMS. Second, we do a numerical and theoretical comparison of the input moment matrices of the LMS and the NLMS. The analysis shows why the performance of the NLMS is less sensitive to the changes in eigenvalue-spread (of the input-correlation matrix) than the LMS.

关键词： Eigenvalues and eigenfunctions signal processing algorithms Sensitivity Random variables Closed-form solutions Standards Technological innovation Adaptive filters LMS NLMS input moment matrices eigenvalue-spread

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Clothing Change Adaptive Person ReID

引用

IEEE signal processing LETTERS 2022年 29卷 304-308页

作者： Zhang, Ziyue Jiang, Shuai Huang, Congzhentao Xu, Richard Yi Da Univ Technol Sydney Broadway Sydney NSW 2007 Australia

Clothing changes and lack of data labels are both crucial challenges in person ReID. For the former challenge, people may occur multiple times wearing different clothes. A majority of current research works for person ReID focuses on benchmarks in which a person's clothing is kept the same all the time. For the latter challenge, most researchers try to transfer information from a labeled source dataset to an unlabeled target dataset. Whereas purely unsupervised training is less used. In this letter, we aim to solve both problems simultaneously. We design a novel unsupervised model, Syn-Person-Cluster ReID, to solve the unlabeled clothing change person ReID problem. We develop a purely unsupervised pipeline equipped with two innovations: synthetic augmentation on person images and feature restriction for the same person. The synthetic augmentation is to supply additional information for the same person, which can then be used as partially supervised inputs to the feature restriction. Extensive experiments on clothing change ReID datasets show the out-performance of our methods.

关键词： Clothing Feature extraction Pipelines Training Cameras Unsupervised learning signal processing algorithms Unsupervised learning clothing change adaptive person image data augmentation person ReID

来源：评论

学校读者我要写书评

暂无评论

Sparsity-Based Time Delay Estimation Through the Matched Filter Outputs

引用

IEEE signal processing LETTERS 2022年 29卷 1769-1773页

作者： Zhang, Yuxuan Jin, Yuxi Wu, Yongqing Hao, Chengpeng Orlando, Danilo Chinese Acad Sci Inst Acoust Beijing 100190 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China Univ Niccolo Cusano I-00166 Rome Italy

In this letter, we deal with the problem of high-resolution time delay estimation (TDE) in multipath environments exploiting the matched filter (MF) outputs data. To this end, we develop a systematic post-processing framework, consisting of two sparsity-based algorithms and a refining procedure aimed at reducing the computational load. The TDE problem is formulated as a sparse signal recovery problem and efficiently solved resorting to a majorization-minimization paradigm and a cyclic procedure. At the design stage, we assume a complex-valued Gaussian distribution model for the MF samples and incorporate a module-product prior that promotes the sparsity more significantly than the conventional complex Laplacian distribution. The preliminary performance assessment, conducted on simulated data, shows that, at least for the considered parameter values, the proposed delay estimators approach the Cramer-Rao bound for different signal-to-noise ratios and bandwidths.

关键词： Estimation signal processing algorithms Delay effects Maximum likelihood estimation Time of arrival estimation signal to noise ratio signal resolution Cyclic estimation high-resolution time delay estimation majorization-minimization multipath environments sparse recovery

来源：评论

学校读者我要写书评

暂无评论

Boundary Effects for EMD-Based algorithms

引用

IEEE signal processing LETTERS 2022年 29卷 1032-1036页

作者： Wang, Yung-Hung Cheng, Shao-Ho Natl Cent Univ Dept Mech Engn Taoyuan 32001 Taiwan

Empirical mode decomposition (EMD) and its improved EMD-based algorithms are adaptive and nonlinear methods that decompose a nonstationary signal into several intrinsic mode functions (IMFs) through the sifting process. When the EMD-based algorithms are applied to decompose a signal polluted with noise in some regions, the noise will spread into the clean region and introduce errors in the IMFs. During the real-time computation of an EMD-based algorithm, the signal is partitioned into a series of overlapping time windows. Points outside the window are discarded, they act as an error source and the error will spread into the window. We will prove that the above two problems are mathematically identical. Due to the complex nature of the sifting iteration, a mathematical theory for the error analysis of the boundary effect is still lacking. Previous studies mainly rely on simulations, and the results show that the error seems to be confined locally near the boundary. Beginning with one sifting iteration, which is the kernel of an EMD-based method, we will theoretically analyze the boundary effects completely based on the sifting process without any approximations, and prove that the error will propagate into the interior domain and may be amplified and thus obscure the meaning of an IMF. Then, we will prove a sufficient condition for the exponential decay of the error. When the error is amplified, we propose a method to resolve this problem. Finally, numerical experiments are conducted to analyze multiple siftings and IMFs.

关键词： signal processing algorithms Mathematical models Indexes Sufficient conditions Splines (mathematics) Real-time systems Partitioning algorithms EMD real-time online noise boundary effect error

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：