12,591 Hits in 6.6 sec

Bayesian Robust Matrix Factorization for Image and Video Processing

Naiyan Wang, Dit-Yan Yeung
2013 2013 IEEE International Conference on Computer Vision  
The proposed methods give competitive experimental results when compared with several state-of-the-art methods on some benchmark image and video processing tasks.  ...  To benefit from the strengths of full Bayesian treatment over point estimation, we propose here a full Bayesian approach to robust matrix factorization.  ...  Bayesian Robust Matrix Factorization In this section, we first present the graphical model and the generative process of our basic Bayesian robust matrix factorization (BRMF) model.  ... 
doi:10.1109/iccv.2013.224 dblp:conf/iccv/WangY13 fatcat:ywcunx4vejbzhismuxqdhljjou

Robust Principal Component Analysis for Background Subtraction: Systematic Evaluation and Comparative Analysis [chapter]

Charles Guyon, Thierry Bouwmans, El-hadi Zahzah
2012 Principal Component Analysis  
et al. (2006)), Incremental Non-negative Matrix Factorization (Bucak et al. (2007) ) and Incremental Rank Tensor (Li et al. (2008) ).  ...  Futur research directions may concern the evalutation of accelerate hardware implementation of robust PCA (Mu et al. (2011) ; Anderson et al. (2011) ) and robust Independent Components Analysis (Yamazaki  ...  For each sequence, the ground truth is provided for twenty images when algorithms have to show their robustness.  ... 
doi:10.5772/38267 fatcat:bkwhvjpfmrbptdsmzkes4kysgy

Bayesian Robust Principal Component Analysis

Xinghao Ding, Lihan He, L. Carin
2011 IEEE Transactions on Image Processing  
A hierarchical Bayesian model is considered for decomposing a matrix into low-rank and sparse components, assuming the observed matrix is a superposition of the two.  ...  For example, in video applications each row (or column) corresponds to a video frame, and we introduce a Markov dependency between consecutive rows in the matrix (corresponding to consecutive frames in  ...  Because of the flexibility of the Bayesian framework, many variants of this matrix factorization form have been proposed to improve the model robustness and performance [26] , [27] .  ... 
doi:10.1109/tip.2011.2156801 pmid:21606026 fatcat:cts7gj6ehjccvejwnqqrlp3pya

Deep Gradient Prior Regularized Robust Video Super-Resolution

Qiang Song, Hangfan Liu
2021 Electronics  
This paper proposes a robust multi-frame video super-resolution (SR) scheme to obtain high SR performance under large upscaling factors.  ...  For robust SR reconstruction, a weighting scheme is proposed to exclude the outlier data.  ...  In this paper, a robust multi-frame video super-resolution scheme is proposed to deal with large upscaling factors.  ... 
doi:10.3390/electronics10141641 fatcat:xp2igllfr5ftbbrbwzswt3w6oi

Information Fusion For Identity Verification

Girija Chetty, Monica Singh
2011 Zenodo  
, and classification using multivariate Bayesian classifiers allows significant improvement in recognition accuracy for low resolution surveillance video scenarios.  ...  for capturing the inherent multimodality in walking gait patterns, and at the same time discriminating the person identity..  ...  Figure 1 shows some of the sample images from the walking video sequences., and Figrue 2 shows multiple frames of the sequences for a person walking in the video clip.  ... 
doi:10.5281/zenodo.1077900 fatcat:rqzohllvhfgbfk5muko3v6lazi

Bayesian Nonparametric Approaches to Abnormality Detection in Video Surveillance

Vu Nguyen, Dinh Phung, Duc-Son Pham, Svetha Venkatesh
2015 Annals of Data Science  
In particular, we employ the Infinite Hidden Markov Model and Bayesian Nonparametric Factor Analysis for stream data segmentation and pattern discovery.  ...  As widely acknowledged in the computer vision community and security management, discovering suspicious events is the key issue for abnormal detection in video surveillance.  ...  Since we follow a part-based decomposition approach for scene understanding, each latent factor is a sparse image having the same dimension of the original video frame.  ... 
doi:10.1007/s40745-015-0030-3 fatcat:56wmpggdgraktipdvg5solnnp4

Robust Online Video Super-Resolution Using an Efficient Alternating Projections Scheme [article]

Ricardo Augusto Borsoi
2020 arXiv   pre-print
reconstructed image quality and robustness.  ...  An accurate and very efficient approximation for the projection operations is also obtained using tools from multidimensional multirate signal processing.  ...  This condition cannot be satisfied when T(z) corresponds to the video SRR system matrix in (7) for any integer decimation factor d > 1.  ... 
arXiv:1909.00073v2 fatcat:text4hdx3fdmpniej2624433ly

Bayesian Robust Tensor Factorization for Incomplete Multiway Data

Qibin Zhao, Guoxu Zhou, Liqing Zhang, Andrzej Cichocki, Shun-Ichi Amari
2016 IEEE Transactions on Neural Networks and Learning Systems  
We propose a generative model for robust tensor factorization in the presence of both missing data and outliers.  ...  For model learning, we develop an efficient closed-form variational inference under a fully Bayesian treatment, which can effectively prevent the overfitting problem and scales linearly with data size.  ...  The Appendix for detailed proofs and derivations and several videos for demonstration are provided in supplementary materials.  ... 
doi:10.1109/tnnls.2015.2423694 pmid:26068876 fatcat:6ur3z46hfreqjhucez7l4vriku

Spatio-temporal coupled Bayesian Robust Principal Component Analysis for road traffic event detection

Shiming Yang, Konstantinos Kalpakis, Alain Biem
2013 16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013)  
We propose a new method for detecting traffic events that impact road traffic conditions by extending the Bayesian Robust Principal Component Analysis (RPCA) approach.  ...  Our proposed method process datastreams in an incremental way with little computational cost, and hence it is suitable to detect events in an online and real-time manner.  ...  We would also like to thank the Minnesota Traffic Observatory for providing us with of raw loop detector dataset, and thank Ms. Yulei Wang and Ms.  ... 
doi:10.1109/itsc.2013.6728263 dblp:conf/itsc/YangKB13 fatcat:6petwaxrwbhp5p2cda7mrsv3ta

Image reconstruction from videos distorted by atmospheric turbulence

Xiang Zhu, Peyman Milanfar, Amir Said, Onur G. Guleryuz
2010 Visual Information Processing and Communication  
) regularization prior, to make the algorithm more robust to noise and estimation error.  ...  To correct geometric distortion and reduce blur in videos that suffer from atmospheric turbulence, a multi-frame image reconstruction approach is proposed in this paper.  ...  requires 10 minutes for the above new method, Bayesian Image Reconstruction The Bayesian reconstruction models the imaging process described in Eq.(1), and the output image is estimated by minimizing  ... 
doi:10.1117/12.840127 dblp:conf/vipc/ZhuM10 fatcat:6vdnzomkljb6dh2lecb2hznhca

Visual Object Tracking Based on 2DPCA and ML

Ming-Xin Jiang, Min Li, Hong-Yu Wang
2013 Mathematical Problems in Engineering  
Compared with other popular methods, our method reduces the computational complexity and is very robust to abnormal changes.  ...  Finally, to further reduce tracking drift, we employ a template update strategy which combines incremental subspace learning and the error matrix.  ...  Acknowledgments This research described in this paper was supported by the Fundamental Research Funds for the Central Universities (DC110321, DC120101132, and DC120101131).  ... 
doi:10.1155/2013/404978 fatcat:hlqtbqa3gjhvzouzwq7vslbdci

Video retrieval using sparse Bayesian reconstruction

Pablo Ruiz, S. Derin Babacan, Li Gao, Zhu Li, Rafael Molina, Aggelos K. Katsaggelos
2011 2011 IEEE International Conference on Multimedia and Expo  
Every day, a huge amount of video data is generated for different purposes and applications. Fast and accurate algorithms for efficient video search and retrieval are therefore essential.  ...  Once the representation (where sparsity is expected) has been chosen and the observations have been taken, the proposed approach utilizes Bayesian modeling and inference to tackle the retrieval problem  ...  ," IEEE Trans. on Image Processing, vol. 11, no. 5, pp. 497-508, May 2002.  ... 
doi:10.1109/icme.2011.6012002 dblp:conf/icmcs/RuizBGLMK11 fatcat:7fisupc63bdrzpmjnbf7khp2xq

Distilling information with super-resolution for video surveillance

Marco Cristani, Dong Seon Cheng, Vittorio Murino, Donato Pannullo
2004 Proceedings of the ACM 2nd international workshop on Video surveillance & sensor networks - VSSN '04  
To overcome these limitations, we propose a novel Bayesian framework based on image super-resolution, that integrates all the informative bits of a target and condenses the redundancy.  ...  We call this process distillation.  ...  These constraints are quite hard and penalizing, in particular for a video sequence, making the super-resolution image estimation possible only in supervised and controlled conditions, strongly reducing  ... 
doi:10.1145/1026799.1026803 fatcat:3t3pks6xtvf2xlor22725g7idu

Super-Resolution Reconstruction of Compressed Video Using Transform-Domain Statistics

B.K. Gunturk, Y. Altunbasak, R.M. Mersereau
2004 IEEE Transactions on Image Processing  
Considerable attention has been directed to the problem of producing high-resolution video and still images from multiple low-resolution images.  ...  Super-resolution techniques that have been designed for raw (i.e., uncompressed) video may not be effective when applied to compressed video because they do not incorporate the compression process into  ...  There are also several Bayesian algorithms that are designed for compressed video.  ... 
doi:10.1109/tip.2003.819221 pmid:15376955 fatcat:6eln5klyf5cxvbix6v4kj2vrfm

Face recognition with independent component-based super-resolution

Osman G. Sezer, Yucel Altunbasak, Aytul Ercil, John G. Apostolopoulos, Amir Said
2006 Visual Communications and Image Processing 2006  
Since face images are high dimensional data which are mostly redundant for the face recognition task, feature extraction methods that reduce the dimension of the data are becoming standard for face analysis  ...  Therefore, we propose new superresolution algorithms using Bayesian estimation and projection onto convex sets methods in feature domain and present a comparative analysis of the proposed algorithms with  ...  Authors also express their gratitude to the members of Computer Vision and Pattern Analysis (VPA) and Telecommunications Laboratories and graduate students in Sabancı University for participating VPA Super-resolution  ... 
doi:10.1117/12.645868 fatcat:b5qoubwdxbfyjb4e6zimmj5c2m
« Previous Showing results 1 — 15 out of 12,591 results