1,099 Hits in 11.2 sec

An approach to partial occlusion using deep metric learning

Chethana Hadya Thammaiah, Trisiladevi Chandrakant Nagavi
2021 International Journal of Informatics and Communication Technology (IJ-ICT)  
Face recognition in forensics is a challenging task due to the presence of partial occlusion features like wearing a hat, sunglasses, scarf, and beard.  ...  Experiments are carried out on disguised faces in the wild (DFW) dataset collected from IIIT Delhi which consists of 1000 subjects in which 600 subjects were used for testing and the remaining 400 subjects  ...  Then approximation algorithm based on greedy approach is designed to perform string matching.  ... 
doi:10.11591/ijict.v10i3.pp204-211 fatcat:wwlwumplzbgmfhliq2sly6i5xi

Illumination robust dictionary-based face recognition

Vishal M. Patel, Tao Wu, Soma Biswas, P. Jonathon Phillips, Rama Chellappa
2011 2011 18th IEEE International Conference on Image Processing  
In this paper, we present a face recognition method based on simultaneous sparse approximations under varying illumination. Our method consists of two main stages.  ...  In the first stage, a dictionary is learned for each face class based on given training examples which minimizes the representation error with a sparseness constraint.  ...  One algorithm for simultaneous sparse approximation based on greedy pursuit, called simultaneous orthogonal matching pursuit (SOMP), was proposed in [7] . SOMP uses a predefined dictionary.  ... 
doi:10.1109/icip.2011.6116670 dblp:conf/icip/PatelWBPC11 fatcat:t5xbqayidfdd5fpz2hup37uq6u

EPML: Expanded Parts Based Metric Learning for Occlusion Robust Face Verification [chapter]

Gaurav Sharma, Frédéric Jurie, Patrick Pérez
2015 Lecture Notes in Computer Science  
We propose a novel Expanded Parts based Metric Learning (EPML) model for face verification.  ...  The model is capable of mining out the discriminative regions at the right locations and scales, for identity based matching of face images.  ...  [25] recently proposed a new partial face recognition approach by aligning partial face patches to holistic gallery faces automatically, hence being robust to occlusions and illumination changes.  ... 
doi:10.1007/978-3-319-16817-3_4 fatcat:2yvqyz6mh5fgxe5j2p5kavprp4


Yongjiao Wang, Chuan Wang, Lei Liang
2015 International Journal on Smart Sensing and Intelligent Systems  
Sparse representation based classification for face images has been one of efficient approaches for face recognition in recent years.  ...  Experimental results show that face recognition method based on sparse representation is comparable to others.  ...  The typical face detection algorithm consists of detection algorithm based on heuristic, detection algorithm based on the detection of skin color region and detection algorithm based on statistical template  ... 
doi:10.21307/ijssis-2017-751 fatcat:pmththpavvbvxou7t254of74iy

Visual tracking and recognition using probabilistic appearance manifolds

Kuang-Chih Lee, Jeffrey Ho, Ming-Hsuan Yang, David Kriegman
2005 Computer Vision and Image Understanding  
Conventional video-based face recognition systems have usually been embodied with two independent components: the tracking and recognition modules.  ...  This paper presents an algorithm for modeling, tracking, and recognizing human faces in video sequences within one integrated framework.  ...  The video sequences used in the experiments as well as a preliminary version of the face recognition system were recorded and implemented during the summer of 2002 at Honda Research Institute.  ... 
doi:10.1016/j.cviu.2005.02.002 fatcat:enr7vvdw4zg4nmlgbki7wbzzj4

Face Recognition Under Occlusions and Variant Expressions With Partial Similarity

Xiaoyang Tan, Songcan Chen, Zhi-Hua Zhou, Jun Liu
2009 IEEE Transactions on Information Forensics and Security  
Recognition in uncontrolled situations is one of the most important bottlenecks for practical face recognition systems.  ...  The effectiveness of the proposed method in handling large expressions, partial occlusions and other distortions is demonstrated on several well-known face databases.  ...  Fig. 8 . 8 Results on partial occlusions on the AR Databases. (a) sunglasses and (b) scarf. Fig. 9 . 9 Examples of FERET face images with simulated occlusions.  ... 
doi:10.1109/tifs.2009.2020772 fatcat:vg4j5asvzrf77evyte37wab4si

Finding Correspondence from Multiple Images via Sparse and Low-Rank Decomposition [chapter]

Zinan Zeng, Tsung-Han Chan, Kui Jia, Dong Xu
2012 Lecture Notes in Computer Science  
A scalable algorithm is proposed to approximate the solution by solving two sub-problems sequentially: minimization of the sum of nuclear norm and l 1 norm for solving relaxed partial permutation matrices  ...  We verify the efficacy and robustness of the proposed method with extensive experiments with both images and videos.  ...  The pioneering works in [15, 17, 18] exploited this property and a greedy algorithm was proposed to recursively find one partial permutation at a time for correspondence.  ... 
doi:10.1007/978-3-642-33715-4_24 fatcat:rstmkbixubgljepryqpulqtpya

Nasal Region-Based 3D Face Recognition under Pose and Expression Variations [chapter]

Hamdi Dibeklioğlu, Berk Gökberk, Lale Akarun
2009 Lecture Notes in Computer Science  
In this work, we propose a fully automatic pose and expression invariant part-based 3D face recognition system. The proposed system is based on pose correction and curvature-based nose segmentation.  ...  We try both one-to-all and Average Nose Model-based methodologies for registration.  ...  Part-based approach is useful to alleviate the pose, facial expression and partial occlusion effects on the recognition performance.  ... 
doi:10.1007/978-3-642-01793-3_32 fatcat:yfxg55baqrdilkafwh2ygkfa2q

Dealing with occlusions in the eigenspace approach

A. Leonardis, H. Bischof
1996 Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
The approach enables us not only to reject outliers and to deal with occlusions but also to simultaneously use multiple classes of eigenimages.  ...  The basic limitations of the current appearance-based matching methods using eigenimages are non-robust estimation of coe cients and inability to cope with problems related to occlusions and segmentation  ...  human face recognition 15, 2] .  ... 
doi:10.1109/cvpr.1996.517111 dblp:conf/cvpr/LeonardisB96 fatcat:lbgchd5yffce7e52sqs5v5ewke

Single/Cross-Camera Multiple-Person Traffic Tracking by Graph Matching

2018 DEStech Transactions on Engineering and Technology Research  
Finally, we propose a new Part-based matching method to get the correlation between different tracklets and apply linear programming and greedy algorithm to handle the data association problem to generate  ...  The experiments on our proposed method demonstrate state-of-the-art performance in multiple person tracking.  ...  At last, discretization is performed to tighten the approximation result for the one-to-one matching.  ... 
doi:10.12783/dtetr/ecame2017/18439 fatcat:f7kcfnn7dfe53g2meiw7qmxoj4

Dictionary-Based Face Recognition Under Variable Lighting and Pose

Vishal M. Patel, Tao Wu, Soma Biswas, P. Jonathon Phillips, Rama Chellappa
2012 IEEE Transactions on Information Forensics and Security  
We present a face recognition algorithm based on simultaneous sparse approximations under varying illumination and pose.  ...  A dictionary is learned for each class based on given training examples which minimizes the representation error with a sparseness constraint.  ...  ACKNOWLEDGMENT The authors would like to thank Fabio Scotti and the anonymous reviewers for their many valuable comments and suggestions which significantly improved this paper.  ... 
doi:10.1109/tifs.2012.2189205 fatcat:hkhexmdvnnewjasn2nkvszym44

Curve normalization for shape retrieval

Nacéra Laiche, Slimane Larabi, Farouk Ladraa, Abdelnour Khadraoui
2014 Signal processing. Image communication  
In this paper, we propose a novel part-based approach for two dimensional (2-D) shape description and recognition.  ...  Finally, based on shape similarity of resulting curves, a shape similarity between an input shape and its reference model is defined. A two-step matching algorithm is proposed.  ...  It can be observed from Fig. 8 that for different degrees of occlusion our algorithm can classify correctly most of the shape queries on occlusion.  ... 
doi:10.1016/j.image.2014.01.009 fatcat:ot2avtfoajbltbg6niegfsnpoq

Hexagonal Descriptor Particle Swarm Optimization with Knowledge-Crowding for Face Recognition

Taqdir ., Renu Dhir
2016 International Journal of Signal Processing, Image Processing and Pattern Recognition  
The proposed methodology is validated with benchmark datasets for analyzing the performance over other methods. based classification based on LNS of the features of the training image.  ...  Automatic identification of various facial expressions with high recognition value is important for human computer interaction as the facial behavior of a human can be treated as an important factor for  ...  ., Regional Campus, Gurdaspur, India and Department of Computer Science & Engineering, in NIT, Jalandhar, India for providing excellent lab facilities that make this work possible.  ... 
doi:10.14257/ijsip.2016.9.9.24 fatcat:2jwvgwhga5hvzaloo257llgave

Action recognition by learnt class-specific overcomplete dictionaries

Tanaya Guha, Rabab K. Ward
2011 Face and Gesture 2011  
This paper presents a sparse signal representation based approach to address the problem of human action recognition in videos.  ...  Given a query sequence, the classifier seeks the dictionary that best approximates the query class. We have evaluated the proposed approach on the standard datasets.  ...  Note that, our method also notably differs from the sparse face recognition approach presented in [8] .  ... 
doi:10.1109/fg.2011.5771388 dblp:conf/fgr/GuhaW11 fatcat:uxbtmlav6baa5dongsu6p34n2u

Sparse representations for image decomposition with occlusions

M. Donahue, D. Geiger, R. Hummel, Tyng-Luh Liu
1996 Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
Our approach is to treat it as a function approximation problem based on an over-redundant basis, and also account for occlusions, where the basis superposition principle is no longer valid.  ...  We are motivated to select a sparse/compact representation of I, and to account for occlusions and noise.  ...  The original matching pursuit is based on the standard L 2 method. In recognition of image with occlusions, the L 2 norm is not suitable.  ... 
doi:10.1109/cvpr.1996.517046 dblp:conf/cvpr/DonahueGLH96 fatcat:7e2vz2u4bzbblfw7kyaksczrli
« Previous Showing results 1 — 15 out of 1,099 results