Filters








1,651 Hits in 6.9 sec

Tensor LRR based subspace clustering

Yifan Fu, Junbin Gao, David Tien, Zhouchen Lin
2014 2014 International Joint Conference on Neural Networks (IJCNN)  
TLRR better captures the global structures of data and provides a robust subspace segmentation from corrupted data.  ...  Subspace clustering groups a set of samples (vectors) into clusters by approximating this set with a mixture of several linear subspaces, so that the samples in the same cluster are drawn from the same  ...  Therefore, it is more reasonable to consider data lying on a mixture of multiple low-dimensional subspaces, with each subspace fitting a subgroup of data.  ... 
doi:10.1109/ijcnn.2014.6889472 dblp:conf/ijcnn/FuGTL14 fatcat:oi6ddrpc5rgotbf7dydlmcsck4

Mixture Matrix Completion [article]

Daniel L. Pimentel-Alarcón
2018 arXiv   pre-print
This paper generalizes these models to what we call mixture matrix completion (MMC): the case where each entry of X corresponds to one of several low-rank matrices.  ...  One typical assumption is that X is low-rank. A more general model assumes that each column of X corresponds to one of several low-rank matrices.  ...  Real Data: MMC for Background Segmentation As discussed in Section 2, robust PCA models a video as the superposition of a low-rank background plus a sparse foreground with no structure.  ... 
arXiv:1808.00616v1 fatcat:b6ljyeckuvflxaakwm2hrc43ve

Robust Subspace Clustering via Half-Quadratic Minimization

Yingya Zhang, Zhenan Sun, Ran He, Tieniu Tan
2013 2013 IEEE International Conference on Computer Vision  
It is a challenging task to learn low-dimensional subspace structures due to the possible errors (e.g., noise and corruptions) existing in high-dimensional data.  ...  The second part aims to maximize the correntropy between a given data point and its low-dimensional representation with other points.  ...  ) method and compare it with stateof-the-art subspace clustering methods, e.g., Local Subspace Analysis (LSA) [27] , Spectral curvature clustering (SCC) [4] , LRR [13] , LSR [17] , Low-Rank Subspace  ... 
doi:10.1109/iccv.2013.384 dblp:conf/iccv/ZhangSHT13 fatcat:qhbmt2o4pzhuvhobiymcxtyle4

Working memory inspired hierarchical video decomposition with transformative representations [article]

Binjie Qin, Haohao Mao, Ruipeng Zhang, Yueqi Zhu, Song Ding, Xu Chen
2022 arXiv   pre-print
Specifically, robust PCA unrolling networks acting as a structure-regularized sensor layer decompose XCA into sparse/low-rank structured representations to separate moving contrast-filled vessels from  ...  decomposed nonlocal patches into orthogonal subspaces for heterogeneous vessel retrieval and interference suppression.  ...  constrained optimization for low-rank subspace updating.  ... 
arXiv:2204.10105v3 fatcat:ifzpeay2qjfvbaznwruwc4dz5m

Subspace segmentation with outliers: A grassmannian approach to the maximum consensus subspace

Nuno Pinho da Silva, Joao Paulo Costeira
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
Besides robustness, it does not rely on prior global detection procedures (e.g., rank of data matrices), which is the case of most current works.  ...  with an efficient optimization algorithm.  ...  We call arbitrary union to a mixture of an unknown number of linear subspaces of unknown dimensions, with arbitrary intersections and containing outliers.  ... 
doi:10.1109/cvpr.2008.4587466 dblp:conf/cvpr/SilvaC08 fatcat:rlqzlj7hlbasvlddgfd5wgmqwy

Union of Low-Rank Subspaces Detector [article]

Mohsen Joneidi, Parvin Ahmadi, Mostafa Sadeghi, Nazanin Rahnavard
2016 arXiv   pre-print
Low-rank assumption for the desired signals implies that the representations of these signals in terms of some proper bases would be sparse.  ...  In this paper, we propose a new detection method based on sparse decomposition in a union of subspaces (UoS) model.  ...  low-rank subspace).  ... 
arXiv:1307.7521v6 fatcat:2bh4owki3zdjdo7gha5zp7aybq

Singular Spectrum Analysis of Gene Expression Profiles of EarlyDrosophila embryo: Exponential-in-Distance Patterns

T. Alexandrov, N. Golyandina, A. Spirov
2008 Research Letters in Signal Processing  
The biological problem under investigation is the decomposition ofbicoidprotein profiles ofDrosophila melanogasterinto the sum of a signal and noise, where the former consists of an exponential-in-distance  ...  However, the subspace spanned by the corresponding eigenvectors does not reflect the finite-rank structure of the trend and is liable to be affected by noise and outliers.  ...  Note that any trend can be approximated by a finite-rank series as the class of finite-rank series includes all kinds of sums of products of polynomials, exponentials and sinusoids.  ... 
doi:10.1155/2008/825758 pmid:21152265 pmcid:PMC2997757 fatcat:v72rg52h5zbgfissw35uvno4ta

Union of low-rank subspaces detector

Mohsen Joneidi, Nazanin Rahnavard, Parvin Ahmadi, Mostafa Sadeghi
2016 IET Signal Processing  
Low-rank assumption for the desired signals implies that the representations of these signals in terms of some proper bases would be sparse.  ...  In this paper, we propose a new detection method based on sparse decomposition in a union of subspaces (UoS) model.  ...  low-rank subspace).  ... 
doi:10.1049/iet-spr.2015.0009 fatcat:g4tiqkf3qvfmje2z3oxtrd4z4u

Tensor LRR and Sparse Coding-Based Subspace Clustering

Yifan Fu, Junbin Gao, David Tien, Zhouchen Lin, Xia Hong
2016 IEEE Transactions on Neural Networks and Learning Systems  
TLRRSC can well capture the global structure and inherent feature information of data and provide a robust subspace segmentation from corrupted data.  ...  Index Terms-Dictionary learning, sparse coding (SC), subspace clustering, tensor low-rank representation (TLRR).  ...  Therefore, we assert that TLRRSC is a noise robust method with low memory usage. VI. CONCLUSION In this paper, we propose TLRR and SC for subspace clustering in this paper.  ... 
doi:10.1109/tnnls.2016.2553155 pmid:27164609 fatcat:rdyohb5qybgxvipppbdv5wepyq

Blind Image Denoising via Dependent Dirichlet Process Tree [article]

Fengyuan Zhu, Guangyong Chen, Jianye Hao, Pheng-Ann Heng
2016 arXiv   pre-print
To model the empirical noise of an image, our method introduces the mixture of Gaussian distribution, which is flexible enough to approximate different continuous distributions.  ...  We apply our method on synthesis and real noisy images with different noise models. Comparing with previous approaches, ours achieves better performance.  ...  This prior introduces the mechanisms of parameter sharing among mixture components.  ... 
arXiv:1601.03117v1 fatcat:cvori2mr3beczm3sd6ut6aumbu

Traditional and recent approaches in background modeling for foreground detection: An overview

Thierry Bouwmans
2014 Computer Science Review  
Then, we conclude with several promising directions for future research.  ...  Robust subspace models The background and the foreground are separated via a robust subspace model which is based on a low-rank and sparse decomposition.  ...  Student's t-mixture model (STMM) has proven to be very robust against noises due to its more heavily-tailed nature compared to Gaussian mixture model [63] but STMM has not been previously applied to  ... 
doi:10.1016/j.cosrev.2014.04.001 fatcat:wccwuwltk5fr7lsgmsu5qbxclm

The algebra and statistics of generalized principal component analysis

Shankar Rao, Harm Derksen, Robert Fossum, Yi Ma, Andrew Wagner, Allen Yang, Chang Wen Chen, Dan Schonfeld, Jiebo Luo
2007 Visual Communications and Image Processing 2007  
We provide a summary of important algebraic properties and statistical facts that are crucial for making GPCA both efficient and robust, even when the given data are corrupted with noise or contaminated  ...  We consider the problem of simultaneously segmenting data samples drawn from multiple linear subspaces and estimating model parameters for those subspaces.  ...  Simulations with Noisy Data We provide a comparison of various algorithms for the estimation and segmentation of subspace arrangements in the presence of noise.  ... 
doi:10.1117/12.707527 dblp:conf/vcip/RaoDFMWY07 fatcat:u3wuo6xylfbepnm5iug5zk2mje

Multiple structure recovery via robust preference analysis

Luca Magri, Andrea Fusiello
2017 Image and Vision Computing  
This paper address the extraction of multiple models from outlier-contaminated data by exploiting preference analysis and low rank approximation.  ...  into many single-model problems, which in turn are tackled with an approach inspired to MSAC (M-estimator SAmple Consensus) coupled with a model-specific scale estimate.  ...  The second level appears in the robust low rank approximation, where Robust PCA and SymNMF are used to gives rise to a soft segmentation where outliers are under-weighted.  ... 
doi:10.1016/j.imavis.2017.09.005 fatcat:ibck4muv2jagvbicz6njlsy76i

Low-Rank Tensor Thresholding Ridge Regression

Kailing Guo, Tong Zhang, Xiangmin Xu, Xiaofen Xing
2019 IEEE Access  
Aiming at preserving the spatial information of tensor data, we incorporate tensor mode-d product with low-rank matrices for selfrepresentation.  ...  INDEX TERMS Tensor, low-rank, subspace clustering. VOLUME 7, 2019 This work is licensed under a Creative Commons Attribution 4.0 License.  ...  We also review noise removal in subspace clustering and acceleration of low-rank methods. A.  ... 
doi:10.1109/access.2019.2944426 fatcat:56fatkxvabbzrchvsdotsjjzp4

A tensor foreground-background separation algorithm based on dynamic dictionary update and active contour detection

Shenglei Pei, Linhao Li, Lijuan Ye, Yongfeng Dong
2020 IEEE Access  
Conventional techniques towards this always consider the background as primary target and tend to adopt low-rank constraint as its estimator, which provides finite (equal to the value of rank) alternatives  ...  Then, the moving foreground is considered as a mixture of active contours and continuous contents.  ...  on calculating the rank of a given video tensor [40] ; Sobral et al. propose a tensor subspace learning scheme to solve high order low-rank model [41] while Li et al. come up with an online solution  ... 
doi:10.1109/access.2020.2992494 fatcat:54mzzckikjdqfgt22nwnn2dvxu
« Previous Showing results 1 — 15 out of 1,651 results