SVM-based face verification with feature set of small size

Lee, Kyung‐Hee; Chung, Yongwha; Byun, Hyeran

doi:10.1049/el:20020591

Cited by 28 publications

(13 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A successful face recognition methodology depends heavily on the particular choice of the features used by the (pattern) classifier [4], [5] .Feature selection in pattern recognition involves the derivation of salient features from the raw input data in order to reduce the amount of data for classification and simultaneously provide prominent information of the original images.Most efforts in the literature have been focused mainly on developing the feature extraction methods [6][7] [8] and employing powerful classifiers such as probabilistic [9] ,Hidden Markov models(HMMS) [10] , neural networks(NNs) [11] [12] and support vector machine(SVM) [13] .…”

Section: ) Appearance-basedwhich Uses Holistic Texture Features Andmentioning

confidence: 99%

Recognition of faces using texture-based principal component analysis and Grassmannian distances analysis

Zhang

2011

SPIE Proceedings

View full text Add to dashboard Cite

This paper introduces a new face recognition method-texture-based Principal Component Analysis (PCA), which employs PCA on texture features.Initially, the eigenspace of texture images is created by eigenvalues and eigenvectors.From this space,the eigentextures are constructed,and most of the eigentextures are selected by using PCA.With these eigentextures, we generalize Grassmannian distances into texture feature space to recognize.We address the problem of face recognition in terms of the subject-specific subspaces instead of image vectors.The proposed method is tested on Essex Face 94 database,and it has been demenstrated to have a promising performance.

show abstract

Section: ) Appearance-basedwhich Uses Holistic Texture Features Andmentioning

confidence: 99%

Recognition of faces using texture-based principal component analysis and Grassmannian distances analysis

Zhang

2011

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…Many researcher claim the advantage of LDA over PCA but Martez et al [19] illustrates why the PCA perform better than the LDA when the training dataset is non data non-uniformly sampled or number of samples per person is small. (3) Classification by employing powerful classifiers such as neural networks [23], [24], support vector machine (SVM) [26], Euclidean distance classifier [21], Mahalanobis distance classifier [22], Hidden Markov Models (HMMs) [25], and extreme learning machine (ELM) [27].…”

Section: Introductionmentioning

confidence: 99%

Performance improvement for 2-D face recognition using multi-classifier and BPN

Abuzneid

Mahmood

2016

2016 IEEE Long Island Systems, Applications and Technology Conference (LISAT)

View full text Add to dashboard Cite

Human face recognition and detection has become a very interesting field for the researcher and this interest is motivated by the huge demand of extensive applications of the real time surveillance system and the static matching system like DMV licenses, port authority and bank system. The image processing, neural network and computer vision are the most area active research areas .Many of papers and approaches have been introduced in the past decades. It is difficult to create and design an efficient computational system for accurate human face recognition because of the complex visual of the human face, which changes dramatically based on the variant effects. In this paper, we propose a new framework for 2D face recognition using five different distance algorithms and Backpropagation Neural network (BPN) to improve conventional Principal Component Analysis (PCA) approach. The performance of the new framework is compared with the performance of the K Nearest-Neighbor (KNN) classifier using five different distance algorithms then we combine them using the least square root algorithm to achieve a higher accuracy. Our experimental results on AT&T (formerly ORL) and Yale face databases show that our proposed method improves the overall performance of face recognition system with respect to existing approaches. The results show the high accuracy can be achieved by using %15 of the image features which increases the recognition performance rate with less computational cost than the existing methods. Index Terms-Principal component analysis (PCA); K Nearest-Neighbor; Backpropagation neural network (BPN).

show abstract

“…All these methods extracts features to optimally represent faces belong to a class and separate the faces. In the literature it has been found that most efforts are given mainly on developing feature extraction methods and employing powerful classifiers such as Euclidean distance classifier [22], neural networks [23], [24], Hidden Markov Models (HMMs) [25], support vector machine (SVM) [26], and extreme learning machine (ELM) [27]. In this paper we don't give the attention on face detection and normalization i.e.…”

Section: Introductionmentioning

confidence: 99%

Performance improvement for face recognition using PCA and two-dimensional PCA

Dandpat

Meher

2013

2013 International Conference on Computer Communication and Informatics

View full text Add to dashboard Cite

Now a days research is going on to design a high performance automatic face recognition system which is really a challenging task for researchers. As faces are complex visual stimuli that differ dramatically, hence developing an efficient computational approach for accurate face recognition is very difficult. In this paper a high performance face recognition algorithm is developed and tested using conventional Principal Component Analysis (PCA) and two dimensional Principal Component Analysis (2DPCA). These statistical transforms are exploited for feature extraction and data reduction. We have proposed here to assign different weight to the only very few nonzero eigenvalues related eigenvectors which are considered as non-trivial principal components for classification. Lastly face recognition task is performed by k-nearest distance measurement. Experimental results on ORL and YALE face databases show that the proposed method improves the performance of face recognition with respect to existing techniques. The results show that better recognition performance can be achieved with less computational cost than that of other existing methods.

show abstract

SVM-based face verification with feature set of small size

Cited by 28 publications

References 3 publications

Recognition of faces using texture-based principal component analysis and Grassmannian distances analysis

Recognition of faces using texture-based principal component analysis and Grassmannian distances analysis

Performance improvement for 2-D face recognition using multi-classifier and BPN

Performance improvement for face recognition using PCA and two-dimensional PCA

Contact Info

Product

Resources

About