Filters








494 Hits in 5.7 sec

Nuclear Norm based Matrix Regression with Applications to Face Recognition with Occlusion and Illumination Changes [article]

Jian Yang, Jianjun Qian, Lei Luo, Fanlong Zhang, Yicheng Gao
2014 arXiv   pre-print
Compared with the current regression methods, the proposed Nuclear Norm based Matrix Regression (NMR) model is more robust for alleviating the effect of illumination, and more intuitive and powerful for  ...  We observe that occlusion and illumination changes generally lead to a low-rank error image.  ...  So, our method is named nuclear norm based matrix regression (NMR).  ... 
arXiv:1405.1207v1 fatcat:3jqdge2tafa5faazdxn3tcfwgy

Robust Face Recognition via Multi-Scale Patch-Based Matrix Regression

Guangwei Gao, Jian Yang, Xiaoyuan Jing, Pu Huang, Juliang Hua, Dong Yue, Zhaohong Deng
2016 PLoS ONE  
However, the recognition performance of nuclear norm-based matrix regression degrades greatly in the face of the small sample size problem.  ...  By making use of the low-rank structural information of the reconstructed error image, the socalled nuclear norm-based matrix regression has been demonstrated to be effective for robust face recognition  ...  Patch-based matrix regression (PMR) To make the model robust and efficient for face recognition with occlusion and illumination changes, matrix regression [29, 30, 32] was proposed using the nuclear  ... 
doi:10.1371/journal.pone.0159945 pmid:27525734 pmcid:PMC4985152 fatcat:mz63uqmmubb7hp3nv74o75zvra

Quaternion matrix regression for color face recognition [article]

Jifei Miao, Kit Ian Kou
2020 arXiv   pre-print
More recently, to better deal with some difficult conditions such as occlusions and illumination, nuclear norm based matrix regression methods have been proposed to characterize the low-rank structure  ...  norm based quaternion matrix regression (NQMR).  ...  CONCLUSIONS Focusing on color face recognition problems, this paper utilizing quaternions to represent the color pixels with RGB channels proposed a nuclear norm based quaternion matrix regression (NQMR  ... 
arXiv:2001.10677v1 fatcat:4kevz7ukxnghdis6vxikzy6z5y

Robust nuclear norm regularized regression for face recognition with occlusion

Jianjun Qian, Lei Luo, Jian Yang, Fanlong Zhang, Zhouchen Lin
2015 Pattern Recognition  
We thus introduce a novel robust nuclear norm regularized regression (RNR) method for face recognition with occlusion.  ...  Based on this point, we propose a nuclear-norm regularized regression model and use the alternating direction method of multipliers (ADMM) to solve it.  ...  proposed a sparse representation based classification (SRC) method to identify human faces with varying illumination changes, occlusion and real disguise [2] .  ... 
doi:10.1016/j.patcog.2015.04.017 fatcat:4z7qi45wmzco7mufio6u3h5qfe

Bilateral Two-Dimensional Matrix Regression Preserving Discriminant Embedding for Corrupted Image Recognition

Jianbo Zhang, Jinkuan Wang, Mingwei Li
2019 IEEE Access  
INDEX TERMS Corrupted image, face recognition, low-rank, matrix regression, nuclear-norm.  ...  Nuclear-norm-based matrix regression (NMR) methods have been successfully applied for the recognition of corrupted images.  ...  ACKNOWLEDGMENT The authors would like to thank the online providers of databases containing faces, including Yale, Extended Yale B, CMU-PIE, AR, and LFW.  ... 
doi:10.1109/access.2019.2892955 fatcat:jv72a4cq5vdkznkvwgm6nlug4m

Robust Low-Rank Regularized Regression for Face Recognition with Occlusion

Jianjun Qian, Jian Yang, Fanglong Zhang, Zhouchen Lin
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops  
We thus introduce a novel robust low-rank regularized regression (RLR 3 ) method for face recognition with occlusion.  ...  Observing that occlusion in a face image generally leads to a low-rank error image, we propose a low-rank regularized regression model and use the alternating direction method of multipliers (ADMM) to  ...  Wright et al. proposed a sparse representation based classification (SRC) method to identify human faces with varying illumination changes, occlusion and real disguise [2] .  ... 
doi:10.1109/cvprw.2014.9 dblp:conf/cvpr/QianYZL14 fatcat:qqlow7wyhbcrll44sxjxtqxuti

Multi-model robust error correction for face recognition

Michael Iliadis, Leonidas Spinoulas, Albert S. Berahas, Haohong Wang, Aggelos K. Katsaggelos
2016 2016 IEEE International Conference on Image Processing (ICIP)  
As such, it proves robust for a range of error inducing factors, such as, varying illumination, occlusion, pixel corruption, disguise or their combinations.  ...  Extensive simulations document the superiority of selecting multiple models for representing the noise term in face recognition problems, allowing the algorithm to achieve near-optimal performance in most  ...  INTRODUCTION Robust error estimation for sparse representation-based classification has been recently investigated in Face Recognition (FR) given frontal views with varying illumination and occlusion as  ... 
doi:10.1109/icip.2016.7532956 dblp:conf/icip/IliadisSBWK16 fatcat:hg5adzzfrvhfzab7fxtlvqhrzi

Learning Low-Rank Class-Specific Dictionary and Sparse Intra-Class Variant Dictionary for Face Recognition

Xin Tang, Guo-can Feng, Xiao-xin Li, Jia-xin Cai, Zhaohong Deng
2015 PLoS ONE  
Face recognition is challenging especially when the images from different persons are similar to each other due to variations in illumination, expression, and occlusion.  ...  Therefore, various robust face recognition techniques have been developed to handle variations in illumination [4] and occlusion [5] .  ...  (a) The accuracy with varying parameters on face recognition without occlusion.  ... 
doi:10.1371/journal.pone.0142403 pmid:26571112 pmcid:PMC4646696 fatcat:6ezdtgtgtjhlzjb7qog2f3c5zm

Low-Rank Laplacian-Uniform Mixed Model for Robust Face Recognition

Jiayu Dong, Huicheng Zheng, Lina Lian
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
For structured errors like continuous occlusion or disguise, we utilize robust nuclear norm to constrain the rank of the error matrix.  ...  Sparse representation based methods have successfully put forward a general framework for robust face recognition through linear reconstruction and sparsity constraints.  ...  Nuclearnorm based matrix regression (NMR) [38] was proposed to approximate a rank function and achieved great performance.  ... 
doi:10.1109/cvpr.2019.01217 dblp:conf/cvpr/DongZL19 fatcat:c4zshqvn2zcfbozn2ccnujpcwe

Low‐rank nonnegative sparse representation and local preservation‐based matrix regression for supervised image feature selection

Xingyu Zhu, Xiuhong Chen
2021 IET Image Processing  
For images with noise or occlusion, the accuracy is improved significantly, up to 4%, which indicates that this method has strong robustness.  ...  Matrix regression has attracted much attention due to directly select some meaningful features from matrix data.  ...  According to the property of matrix rank and the definition of the nuclear norm, the problem ( 5 ) can be relaxed into the following convex optimization problem by replacing the rank with nuclear norm  ... 
doi:10.1049/ipr2.12281 fatcat:tt6co6nvpzfqfj7yr4xambc7iu

Discriminant Incoherent Component Analysis

Christos Georgakis, Yannis Panagakis, Maja Pantic
2016 IEEE Transactions on Image Processing  
To this end, a suitable optimization problem, involving the minimization of nuclear-and 1-norm, is solved.  ...  Emphasis is placed on face analysis tasks, namely joint face and expression recognition, face recognition under varying percentages of training data corruption, subject-independent expression recognition  ...  The face images exhibit variations with respect to expression, illumination and two types of occlusion, that is, sunglasses and scarf (see Fig. 3 ).  ... 
doi:10.1109/tip.2016.2539502 pmid:27008268 fatcat:tmmlmwm7yjfatpm2a53gl7xrhm

Face Recognition via Compact Second-Order Image Gradient Orientations

He-Feng Yin, Xiao-Jun Wu, Cong Hu, Xiaoning Song
2022 Mathematics  
These results indicate that the proposed method is superior to its competing approaches with few training samples, and even outperforms some prevailing deep-neural-network-based approaches.  ...  CSOIGO is evaluated under real-world disguise, synthesized occlusion, and mixed variations.  ...  [7] presented nuclear norm-based matrix regression (NMR), which employs two dimensional image-matrix-based error model rather than the one-dimensional pixel-based error model.  ... 
doi:10.3390/math10152587 fatcat:ymvqyhjd5fcb3iaesw5u5jq43q

Face recognition via compact second order image gradient orientations [article]

He-Feng Yin, Xiao-Jun Wu, Xiaoning Song
2022 arXiv   pre-print
Experimental results indicate that the proposed method is superior to its competing approaches with few training samples, and even outperforms some prevailing deep neural network based approaches.  ...  CSOIGO is evaluated under real-world disguise, synthesized occlusion and mixed variations.  ...  Yang et al. 14 presented nuclear norm based matrix regression (NMR) which employs two dimensional image-matrix-based error model rather than the one dimensional pixel-based error model.  ... 
arXiv:2201.09246v2 fatcat:l3s2zqr3mrbwdkaz44awzqrhm4

A Unified Weight Learning and Low-Rank Regression Model for Robust Complex Error Modeling [article]

Miaohua Zhang, Yongsheng Gao, Jun Zhou
2020 arXiv   pre-print
One of the most important problems in regression-based error model is modeling the complex representation error caused by various corruptions and environment changes in images.  ...  For example, in robust face recognition, images are often affected by varying types and levels of corruptions, such as random pixel corruptions, block occlusions, or disguises.  ...  Since the proposed method is a regression-based model, we test and compare it with 8 recently published regression-based face recognition approaches, including RRC-L1 and RRC-L2 [8], HQ-A and HQ-M [35]  ... 
arXiv:2005.04619v4 fatcat:5asvj7gnjbdknjnyxudr5arer4

稳健人脸感知方法在人体测温系统中的应用

梦凯 闫, 健 杨, 以成 高, 建军 钱, 曦 程
2020 Scientia Sinica Informationis  
Nuclear norm based matrix regression with applications to face recognition with occlusion and illumination changes.  ...  Robust nuclear norm-based matrix regression with applications to robust face recognition. IEEE Trans Image Process, 2017, 26: 2286-2295 4 Luo L, Yang J, Zhang B, et al.  ... 
doi:10.1360/ssi-2020-0047 fatcat:tnelkqfqarawte4gqdznmrfjm4
« Previous Showing results 1 — 15 out of 494 results