scholarly journals How Does Gender Balance In Training Data Affect Face Recognition Accuracy?

2020 ◽  
Author(s):  
Vitor Albiero ◽  
Kai Zhang ◽  
Kevin W. Bowyer
Keyword(s):  
Face Recognition ◽  
Recognition Accuracy ◽  
Training Data ◽  
Gender Balance

scholarly journals Generating photo-realistic training data to improve face recognition accuracy

2021 ◽  
Vol 134 ◽  
pp. 86-94
Author(s):  
Daniel Sáez Trigueros ◽  
Li Meng ◽  
Margaret Hartnett
Keyword(s):  
Face Recognition ◽  
Recognition Accuracy ◽  
Training Data

scholarly journals Two Factors in Face Recognition: Whether You Know the Person’s Face and Whether You Share the Person’s Race

Perception ◽  
2021 ◽  
pp. 030100662110140
Author(s):  
Xingchen Zhou ◽  
A. M. Burton ◽  
Rob Jenkins
Keyword(s):  
Face Recognition ◽  
Factorial Design ◽  
Performance Measures ◽  
Recognition Accuracy ◽  
The Other ◽  
Familiarity Effect ◽  
Race Effect ◽  
Two Factors ◽  
Unfamiliar Face

One of the best-known phenomena in face recognition is the other-race effect, the observation that own-race faces are better remembered than other-race faces. However, previous studies have not put the magnitude of other-race effect in the context of other influences on face recognition. Here, we compared the effects of (a) a race manipulation (own-race/other-race face) and (b) a familiarity manipulation (familiar/unfamiliar face) in a 2 × 2 factorial design. We found that the familiarity effect was several times larger than the race effect in all performance measures. However, participants expected race to have a larger effect on others than it actually did. Face recognition accuracy depends much more on whether you know the person’s face than whether you share the same race.

Navon Letters Affect Face Learning and Face Retrieval

2009 ◽  
Vol 56 (4) ◽  
pp. 258-264 ◽  
Author(s):  
Michael B. Lewis ◽  
Claire Mills ◽  
Peter J. Hills ◽  
Nicola Weston
Keyword(s):  
Face Recognition ◽  
Present Experiment ◽  
Recognition Accuracy ◽  
Underlying Mechanism ◽  
Performance Accuracy ◽  
Face Retrieval ◽  
Large Letter

Identifying the local letters of a Navon letter (a large letter made up of smaller different letters) prior to recognition causes impairment in accuracy, while identifying the global letters of a Navon letter causes an enhancement in recognition accuracy ( Macrae & Lewis, 2002 ). This effect may result from a transfer-inappropriate processing shift (TIPS) ( Schooler, 2002 ). The present experiment extends research on the underlying mechanism of this effect by exploring this Navon effect on face learning as well as face recognition. The results of the two experiments revealed that when the Navon task used at retrieval was the same as that used at encoding then the performance accuracy is enhanced, whereas when the processing operations mismatch at retrieval and at encoding, this impairs recognition accuracy. These results provide support for the TIPS explanation of the Navon effect.

A TIED-MIXTURE 2D HMM FACIAL IMAGE RETRIEVAL SYSTEM

2004 ◽  
Vol 13 (05) ◽  
pp. 1133-1146
Author(s):  
H. OTHMAN ◽  
T. ABOULNASR
Keyword(s):  
Face Recognition ◽  
Recognition System ◽  
Training Data ◽  
Expected Improvement ◽  
Model Case ◽  
Face Recognition System ◽  
Image Retrieval System ◽  
The Face ◽  
Institute Of Technology ◽  
Observation Space

In this paper, the effect of mixture tying on a second-order 2D Hidden Markov Model (HMM) is studied as applied to the face recognition problem. While tying HMM parameters is a well-known solution in the case of insufficient training data that leads to nonrobust estimation, it is used here to improve the overall performance in the small model case where the resolution in the observation space is the main problem. The fully-tied-mixture 2D HMM-based face recognition system is applied to the facial database of AT&T and the facial database of Georgia Institute of Technology. The performance of the proposed 2D HMM tied-mixture system is studied and the expected improvement is confirmed.

Low-Resolution Face Recognition of Multi-Scale Blocking CS-LBP and Weighted PCA

2016 ◽  
Vol 30 (08) ◽  
pp. 1656005 ◽  
Author(s):  
Jiadi Li ◽  
Zhenxue Chen ◽  
Chengyun Liu
Keyword(s):  
Face Recognition ◽  
Local Binary Pattern ◽  
Recognition Accuracy ◽  
Recognition Rate ◽  
Principal Component ◽  
Low Resolution ◽  
Multi Scale ◽  
Face Features ◽  
Weighted Principal Component ◽  
Pca Algorithm

A novel method is proposed in this paper to improve the recognition accuracy of Local Binary Pattern (LBP) on low-resolution face recognition. More precise descriptors and effectively face features can be extracted by combining multi-scale blocking center symmetric local binary pattern (CS-LBP) based on Gaussian pyramids and weighted principal component analysis (PCA) on low-resolution condition. Firstly, the features statistical histograms of face images are calculated by multi-scale blocking CS-LBP operator. Secondly, the stronger classification and lower dimension features can be got by applying weighted PCA algorithm. Finally, the different classifiers are used to select the optimal classification categories of low-resolution face set and calculate the recognition rate. The results in the ORL human face databases show that recognition rate can get 89.38% when the resolution of face image drops to 12[Formula: see text]10 pixel and basically satisfy the practical requirements of recognition. The further comparison of other descriptors and experiments from videos proved that the novel algorithm can improve recognition accuracy.

SUBSPACE LEARNING BASED ON LAPLACIAN EIGENMAPS AND LDA FOR FACE RECOGNITION

2006 ◽  
Vol 03 (01) ◽  
pp. 45-51
Author(s):  
YANWEI PANG ◽  
ZHENGKAI LIU ◽  
YUEFANG SUN
Keyword(s):  
Face Recognition ◽  
Subspace Learning ◽  
Dimensional Subspace ◽  
Training Data ◽  
Objective Functions ◽  
Recognition Method ◽  
Laplacian Eigenmaps ◽  
Linear Discriminant ◽  
Structure Information ◽  
Low Dimensional

Subspace-based face recognition method aims to find a low-dimensional subspace of face appearance embedded in a high-dimensional image space. The differences between different methods lie in their different motivations and objective functions. The objective function of the proposed method is formed by combining the ideas of linear Laplacian eigenmaps and linear discriminant analysis. The actual computation of the subspace reduces to a maximum eigenvalue problem. Major advantage of the proposed method over traditional methods is that it utilizes both local manifold structure information and discriminant information of the training data. Experimental results on the AR face databases demonstrate the effectiveness of the proposed method.

The Comparison of Laplacianfaces QR Decomposition and Linear Discriminant Analysis QR Decomposition Algorithm for Face Recognition System on Orthogonal Subspace

2014 ◽  
Vol 71 (1) ◽  
Author(s):  
Purbandini Purbandini
Keyword(s):  
Face Recognition ◽  
Discriminant Analysis ◽  
Linear Discriminant Analysis ◽  
Selection Process ◽  
Recognition System ◽  
Qr Decomposition ◽  
Training Data ◽  
Distance Method ◽  
Linear Discriminant ◽  
Face Recognition System

Development of an optimal face recognition system will greatly depend on the characteristics of the selection process are as a basis to pattern recognition. In the characteristic selection process, there are 2 aspects that will be of mutual influence such the reduction of the amount of data used in the classification aspects and increasing discrimination ability aspects. Linear Discriminat Analysis method helps presenting the global structure while Laplacianfaces method is one method that is based on appearance (appearance-based method) in face recognition, in which the local manifold structure presented in the adjacency graph mapped from the training data points. Linear Discriminant Analysis QR decomposition has a computationally low cost because it has small dimensions so that the efficiency and scalability are very high when compared with algorithms of other Linear Discriminant Analysis methods. Laplacianfaces QR decomposition was a algorithm to obtain highly speed and accuracy, and tiny space to keep data on the face recognition. This algorithm consists of 2 stages. The first stage maximizes the distance of between-class scatter matrices by using QR decomposition and the second stage to minimize the distance of within-class scatter matrices. Therefore, it is obtained an optimal discriminant in the data. In this research, classification using the Euclidean distance method. In these experiments using face databases of the Olivetti-Att-ORL, Bern and Yale. The minimum error was achieved with the Laplacianfaces QR decomposition and Linear Discriminant Analysis QR decomposition are 5.88% and 9.08% respectively. 

scholarly journals Analysis of Gender Inequality In Face Recognition Accuracy

2020 ◽  
Author(s):  
Vitor Albiero ◽  
K.S. Krishnapriya ◽  
Kushal Vangara ◽  
Kai Zhang ◽  
Michael C. King ◽  
...  
Keyword(s):  
Face Recognition ◽  
Gender Inequality ◽  
Recognition Accuracy
Sign in / Sign up

Export Citation Format

Share Document