Filters








9,480 Hits in 7.4 sec

Sparse Graph Based Deep Learning Networks for Face Recognition

Renjie WU, Sei-ichiro KAMATA
2018 IEICE transactions on information and systems  
This paper devoted to the story explain of two properties of our graph -sparse and depth. Sparse can be advantageous since features are more likely to be linearly separable and they are more robust.  ...  The proposed method achieves high recognition rates of 99.61% (94.67%) on the benchmark LFW (YTF) facial evaluation database. key words: face recognition, atom decomposition, sparse graph reconstruction  ...  And then, these 6,000 face pairs are tested for face recognition. For this experiment, all the captured face can be directly obtained by traditional face detection method.  ... 
doi:10.1587/transinf.2017pcp0012 fatcat:pwj2t5b5f5h2jk32dufnalsn34

A Survey of Sparse Representation: Algorithms and Applications

Zheng Zhang, Yong Xu, Jian Yang, Xuelong Li, David Zhang
2015 IEEE Access  
The main purpose of this article is to provide a comprehensive study and an updated review on sparse representation and to supply a guidance for researchers.  ...  Sparse representation also has a good reputation in both theoretical research and practical applications. Many different algorithms have been proposed for sparse representation.  ...  They would also like to thank Dr. Zhihui Lai, Dr. Jinxing Liu and Xiaozhao Fang for constructive suggestions.  ... 
doi:10.1109/access.2015.2430359 fatcat:fdi57s5xxfc3jekrgbgxigkt2q

Multipath Sparse Coding Using Hierarchical Matching Pursuit

Liefeng Bo, Xiaofeng Ren, Dieter Fox
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
We propose Multipath Hierarchical Matching Pursuit (M-HMP), a novel feature learning architecture that combines a collection of hierarchical sparse features for image classification to capture multiple  ...  While progress in deep learning shows the importance of learning features through multiple layers, it is equally important to learn features through multiple paths.  ...  Acknowledgments This work was funded in part by the Intel Science and Technology Center for Pervasive Computing and by ONR MURI grant N00014-07-1-0749.  ... 
doi:10.1109/cvpr.2013.91 dblp:conf/cvpr/BoRF13 fatcat:jepvukvz5jc2jk5ov7hvd6tiau

A unified SWSI–KAMs framework and performance evaluation on face recognition

Songcan Chen, Lei Chen, Zhi-Hua Zhou
2005 Neurocomputing  
In the end, the SWSI-KAM adopting Exponential kernel with different connectivities was emphatically investigated for robustness based on those face images which are added random noises and/or partially  ...  performance is almost as well as, even better than, corresponding KAMs with full connectivity.  ...  Acknowledgement We thank anonymous reviewers very much for their valuable comments and suggestions for improving presentation of this paper. And at the same time we also thank National Science  ... 
doi:10.1016/j.neucom.2005.02.001 fatcat:kg7eoqaj2fhljo4oebjfkyedka

Riemannian Dictionary Learning and Sparse Coding for Positive Definite Matrices [article]

Anoop Cherian, Suvrit Sra
2015 arXiv   pre-print
Inspired by the great success of dictionary learning and sparse coding for vector-valued data, our goal in this paper is to represent data in the form of SPD matrices as sparse conic combinations of SPD  ...  To that end, we formulate a novel Riemannian optimization objective for dictionary learning and sparse coding in which the representation loss is characterized via the affine invariant Riemannian metric  ...  Dataset: In this experiment, we evaluate the performance of the Riemannian DLSC setup to deal with a larger dataset of high-dimensional covariance descriptors for face recognition.  ... 
arXiv:1507.02772v2 fatcat:rbtwzfrsjrahzp3wdyvw277fma

An Efficient Two-Stage Sparse Representation Method

Chengyu Peng, Hong Cheng, Manchor Ko
2016 International journal of pattern recognition and artificial intelligence  
We apply sparse coding to the signals on the dictionary in the first stage, and obtain the training and testing coefficients respectively.  ...  We propose a new method called Two-Stage Sparse Representation (TSSR) to tackle this problem.  ...  Apply sparse coding to obtain representing coefficients x ∈ R N over Ω.  ... 
doi:10.1142/s0218001416510010 fatcat:ea2icp7ckzavjdl5gs5kcsue64

Multi-Person Recognition Using Viola-Jones and Kalman Filter for Target Tracking

2017 International Journal of Science and Research (IJSR)  
The first phase detects all the faces in the crowded scenes and second phase detects the face of the target person using Viola-Jones algorithm for face recognition and third phase tracks the target person  ...  This paper presents a multi-level framework for target tracking in simple and complex environments. Kalman filter is using to perform target tracking.  ...  The coding coefficients of the blocks spanning multiple templates are integrated by averaging and alignment pooling to obtain a robust representation of the target object.  ... 
doi:10.21275/art20175195 fatcat:tvpjmvwsyzgmhdtpqxvb4gu7zu

Robust Visual Tracking via Collaborative Voting with Structured Sparse Representation

Yang Liu, Yibo Li, Xiaofei Ji, Yangyang Wang
2017 International Journal of Hybrid Information Technology  
Sparse representation based methods have recently attracted much attention in visual tracking due to the robustness to corruption, occlusion and other challenging issues.  ...  Different from previous methods, visual tracking is formulated as an object recognition problem in the proposed method, which makes the tracking task more robust to occlusion.  ...  Motivated by the successful application of sparse representation in face recognition, we consider tracking task as recognition problem, in which templates and candidates are used as observed samples and  ... 
doi:10.14257/ijhit.2017.10.1.11 fatcat:xroa5tgr6jelfbj6y2xsfy7zpm

Face Recognition System under Varying Lighting Conditions

P. Kalaiselvi
2013 IOSR Journal of Computer Engineering  
Making recognition more reliable under uncontrolled lighting conditions is one of the most important challenges for practical face recognition systems.  ...  We further increase robustness by introducing Phase Congruency. The resulting method provides a face verification rate of 88.1% at 0.1% false accept rate.  ...  Fig. 2 . 1 . 21 Stages of full face recognition method Face Recognition System under Varying Lighting Conditions www.iosrjournals.org Fig. 3 3 Fig. 3.1.  ... 
doi:10.9790/0661-1437988 fatcat:2v6szl7u3bcuhhkcredz67dfmy

Salient views and view-dependent dictionaries for object recognition

Yi-Chen Chen, Vishal M. Patel, Rama Chellappa, P. Jonathon Phillips
2015 Pattern Recognition  
Furthermore, to evaluate our method, we introduce the notion of view-dependent dictionaries built from salient views for applications in 3D object recognition and retrieval.  ...  Keywords: Salient view, characteristic view class, view geometry, sparse representation, view-dependent dictionaries, object recognition.  ...  For each side view class, we pick only one view with the minimum sparse-to-full reconstruction error (i.e., l 1 = 1).  ... 
doi:10.1016/j.patcog.2015.01.013 fatcat:hnubmd3f6je4tovevaceuhzfbi

Towards Reading Beyond Faces for Sparsity-Aware 3D/4D Affect Recognition

Muzammil Behzad, Nhat Vo, Xiaobai Li, Guoying Zhao
2021 Neurocomputing  
We first propose a novel augmentation method to combat the data limitation problem for deep learning, specifically given 3D/4D face meshes.  ...  For training, the TOP-landmarks and sparse representations are used to train a long short-term memory (LSTM) network for 4D data, and a pre-trained network for 3D data.  ...  Lastly, the authors wish to acknowledge CSC -IT Center for Science, Finland, for computational resources.  ... 
doi:10.1016/j.neucom.2021.06.023 fatcat:gactyi4m3fbk5jywbtc4xrq7wm

3D HMM-based Facial Expression Recognition using Histogram of Oriented Optical Flow

Sheng H Kung, Mohamed A. Zohdy, Djamel Bouchaffra
2015 Transactions on Machine Learning and Artificial Intelligence  
Clearly, HCI can lead to vast improvement in the quality of life for humans. Facial expression and emotion recognition will raise the compassion level of HCI application.  ...  We propose a 3D Hidden Markov Model (HMM) approach to recognizing human facial expressions and associated emotions. To the best of our knowledge, this is the first application of 3D  ...  ACKNOWLEDGEMENT We thank Jasser Jasser of ECE Department of Oakland University for assistance with development and running of the experiment and sharing of his pearls of wisdom with us during the course  ... 
doi:10.14738/tmlai.36.1661 fatcat:2xe2jcwa2ffy7buu5ywh6jvq6i

Regular Partitions and Their Use in Structural Pattern Recognition [article]

Marco Fiorucci
2020 arXiv   pre-print
We first extend an heuristic version of the RL to improve its efficiency and its robustness. We use the proposed algorithm to address graph-based clustering and image segmentation tasks.  ...  This high-throughput generation calls for the development of new effective methods to store, retrieve, understand and process massive network data.  ...  to deal with sparse graphs.  ... 
arXiv:1909.07420v2 fatcat:yconxoyh4ffezb6ahglrdbnruy

Analyzing trajectories on Grassmann manifold for early emotion detection from depth videos

Taleb Alashkar, Boulbaba Ben Amor, Stefano Berretti, Mohamed Daoudi
2015 2015 11th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (FG)  
Two different representations have been proposed to address the problems of face recognition and emotion detection.  ...  They are respectively (1) a dictionary (of subspaces) representation associated to Dictionary Learning and Sparse Coding techniques and (2) a time-parameterized curve (trajectory) representation on the  ...  the sample signal to be coded, D is a dictionary (a n × N matrix being N the number of training samples) with atoms D i ∈ R n in its columns, and λ the sparse regularization parameter.  ... 
doi:10.1109/fg.2015.7163122 dblp:conf/fgr/AlashkarABD15 fatcat:53lkhk6apzaqndylmu5iphqupe

Unconstrained Biometric Recognition: Summary of Recent SOCIA Lab. Research [article]

Varsha Balakrishnan
2020 arXiv   pre-print
The idea is that it can be used as basis for someone wishing to entering in this research topic.  ...  obtaining such extremely ambitious kind of automata.  ...  In this setting, identification is regarded as a variable selection and regularization problem, with sparse linear regression techniques being used to infer the matching probability with respect to each  ... 
arXiv:2001.09703v2 fatcat:hugkig4wxvgaldscwbobn6yhuy
« Previous Showing results 1 — 15 out of 9,480 results