Is gender classification across ethnicity feasible using discriminant functions

doi:10.1109/IJCB.2011.6117524

Home
/
Papers
/
Is gender classification across ethnicity feasible using discriminant functions

Proceedings Article•DOI•

Is gender classification across ethnicity feasible using discriminant functions

Tejas I. Dhamecha¹, Anush Sankaran¹, Richa Singh¹, Mayank Vatsa¹•Institutions (1)

Indraprastha Institute of Information Technology¹

11 Oct 2011-International Journal of Central Banking (IEEE)-pp 1-7

TL;DR: The results suggest that linear discriminant functions provide good generalization capability even with limited number of training samples, principal components and with cross-ethnicity variations.

read less

Abstract: Over the years, automatic gender recognition has been used in many applications. However, limited research has been done on analyzing gender recognition across ethnicity scenario. This research aims at studying the performance of discriminant functions including Principal Component Analysis, Linear Discriminant Analysis and Subclass Discriminant Analysis with the availability of limited training database and unseen ethnicity variations. The experiments are performed on a heterogeneous database of 8112 images that includes variations in illumination, expression, minor pose and ethnicity. Contrary to existing literature, the results show that PCA provides comparable but slightly better performance compared to PCA+LDA, PCA+SDA and PCA+SVM. The results also suggest that linear discriminant functions provide good generalization capability even with limited number of training samples, principal components and with cross-ethnicity variations.

...read moreread less

Citations

PDF

Open Access

More filters

Journal Article•DOI•

Recent advances in facial soft biometrics

[...]

Olasimbo Ayodeji Arigbabu¹, Sharifah Mumtazah Syed Ahmad¹, Wan Azizun Wan Adnan¹, Salman Yussof²•Institutions (2)

Universiti Putra Malaysia¹, Universiti Tenaga Nasional²

01 May 2015-The Visual Computer

TL;DR: This paper presents state-of-the-art techniques in facial soft biometrics research by describing the type of traits, feature extraction methods, and the application domains, and indicates the most recent and valuable results attained.

...read moreread less

Abstract: Face as a biometric attribute has been extensively studied over the past few decades. Even though, satisfactory results are already achieved in controlled environments, the practicality of face recognition in realistic scenarios is still limited by several challenges, such as, expression, pose, occlusion, etc. Recently, the research direction is concentrating on the prospects of complementing face recognition systems with facial soft biometric traits. The ease of extracting facial soft biometrics under several varying conditions has mainly resulted in the ability of using the traits to, either improve the performance of traditional face recognition systems, or performing recognition solely based on many facial soft biometrics. This paper presents state-of-the-art techniques in facial soft biometrics research by describing the type of traits, feature extraction methods, and the application domains. It indicates the most recent and valuable results attained, while also highlighting some possible future scientific research directions to be investigated.

...read moreread less

28 citations

Book Chapter•DOI•

Facial Ethnic Appearance Synthesis

[...]

Felix Juefei-Xu¹, Marios Savvides¹•Institutions (1)

Carnegie Mellon University¹

06 Sep 2014

TL;DR: This work has explored several subspace reconstruction methods for facial ethnic appearance synthesis and proposed dual subspace modeling using the Fukunaga Koontz transform (FKT) yields much better facial ethnic synthesis results than the \(\ell _1\) minimization and the principal component analysis (PCA) reconstruction method.

...read moreread less

Abstract: In this work, we have explored several subspace reconstruction methods for facial ethnic appearance synthesis (FEAS). In our experiments, our proposed dual subspace modeling using the Fukunaga Koontz transform (FKT) yields much better facial ethnic synthesis results than the \(\ell _1\) minimization, the \(\ell _2\) minimization and the principal component analysis (PCA) reconstruction method. With that, we are able to automatically and efficiently synthesize different facial ethnic appearance and alter the facial ethnic appearance of the query image to any other ethnic appearance as desired. Our technique well preserves the facial structure of the query image and simultaneously synthesize the skin tone and ethnic features that best matches target ethnicity group. Facial ethnic appearance synthesis can be applied to synthesizing facial images of a particular ethnicity group for unbalanced database, and can be used to train ethnicity invariant classifiers by generating multiple ethnic appearances of the same subject in the training stage.

...read moreread less

18 citations

Cites background from "Is gender classification across eth..."

...For ethnicity classification [36, 5, 26, 7, 34, 1], it is crucial that the researchers obtain a balanced database with subjects from all ethnic groups(1) equally pre1 As is commonly adopted in the literature, the classification of ethnicity boils down to the 3-class case (asian, black and white), or the 4-class case (east asian, south asian,...
[...]

Proceedings Article•DOI•

Soft biometric trait classification from real-world face videos conditioned on head pose estimation

[...]

Meltem Demirkus¹, Doina Precup¹, James J. Clark¹, Tal Arbel¹•Institutions (1)

McGill University¹

16 Jun 2012

TL;DR: A temporal, probabilistic framework is proposed first to robustly estimate continuous head pose angles from real-world videos, and then to decide on the appropriate set of frames and features to use in a temporal fusion scheme for soft biometric trait classification.

...read moreread less

Abstract: Recently, soft biometric trait classification has been receiving more attention in the computer vision community due to its wide range of possible application areas. Most approaches in the literature have focused on trait classification in controlled environments, due to the challenges presented by real-world environments, i.e. arbitrary facial expressions, arbitrary partial occlusions, arbitrary and nonuniform illumination conditions and arbitrary background clutter. In recent years, trait classification has started to be applied to real-world environments, with some success. However, the focus has been on estimation from single images or video frames, without leveraging the temporal information available in the entire video sequence. In addition, a fixed set of features are usually used for trait classification without any consideration of possible changes in the facial features due to head pose changes. In this paper, we propose a temporal, probabilistic framework first to robustly estimate continuous head pose angles from real-world videos, and then use this pose estimate to decide on the appropriate set of frames and features to use in a temporal fusion scheme for soft biometric trait classification. Experiments performed on large, real-world video sequences show that our head pose estimator outperforms the current state-of-the-art head pose approaches (by up to 51%), whereas our head pose conditioned biometric trait classifier (for the case of gender classification) outperforms the current state-of-the-art approaches (by up to 31%).

...read moreread less

16 citations

Cites background from "Is gender classification across eth..."

...Despite the wide literature on soft biometric trait classification [30, 22, 9, 24, 13, 28, 4, 16, 32, 29, 6, 23, 5, 10, 19, 33] and head pose estimation [25, 31, 1, 26, 3, 8], most of these approaches are not built for unconstrained environments (see Section 2 for details)....
[...]

Journal Article•DOI•

Gender Classification using Facial Embeddings: A Novel Approach

[...]

Avinash Swaminathan¹, Mridul Chaba¹, Deepak Kumar Sharma¹, Yogesh Chaba²•Institutions (2)

Netaji Subhas Institute of Technology¹, Guru Jambheshwar University of Science and Technology²

01 Jan 2020-Procedia Computer Science

TL;DR: A novel method to predict Gender of a person by applying various Machine Learning Classification Techniques on Facial Em-beddings has been proposed and the maximum accuracy obtained is 97%.

...read moreread less

15 citations

Dissertation•

Gender Classification from Facial Images using PCA and SVM

[...]

Pankaj Kumar Sa

01 Jan 2013

TL;DR: An investigation has been made on gender classification through facial images using principal component analysis (PCA), and support vector machine (SVM), which is a dimensionality reduction technique, which is used to represent each image as a feature vector in a low dimensional subspace.

...read moreread less

Abstract: Biometrics is the use of physical characteristics like face, fingerprints, iris etc of an individual for personal identification Some of the challenging problems of face biometrics are face detection, face recognition, and face identification These problems are being researched by the computer vision community for the last few decades Considering the large population, the authentication process of an individual usually consumes a significant amount of time One of the possible solutions is to divide the population into two halves based on gender This will help to reduce the search space of authentication to almost half of the existing data and save substantial amount of time Gender identification through face demands use of strong discriminative features and robust classifiers to separate the female and male faces without any ambiguity In this thesis, an investigation has been made on gender classification through facial images using principal component analysis (PCA), and support vector machine (SVM) PCA is a dimensionality reduction technique, which is used to represent each image as a feature vector in a low dimensional subspace SVM is a binary classifier for which PCA is the input in the form of features and predicts which of the two possible classes forms the output Initially face region is extracted using a proposed skin colour segmentation approach The face region is then subjected to PCA for feature extraction, which encodes second order statistics of data These principal components are fed as input to SVM for classification

...read moreread less

6 citations

References

PDF

Open Access

More filters

Book Chapter•DOI•

I and J

[...]

William Marsden

01 Jan 2012

139,059 citations

Statistical learning theory

[...]

Vladimir Vapnik

01 Jan 1998

TL;DR: Presenting a method for determining the necessary and sufficient conditions for consistency of learning process, the author covers function estimates from small data pools, applying these estimations to real-life problems, and much more.

...read moreread less

Abstract: A comprehensive look at learning and generalization theory. The statistical theory of learning and generalization concerns the problem of choosing desired functions on the basis of empirical data. Highly applicable to a variety of computer science and robotics fields, this book offers lucid coverage of the theory as a whole. Presenting a method for determining the necessary and sufficient conditions for consistency of learning process, the author covers function estimates from small data pools, applying these estimations to real-life problems, and much more.

...read moreread less

26,531 citations

"Is gender classification across eth..." refers result in this paper

...It is interesting to note that the maximum performance of PCA is higher than the best results obtained by PCA+LDA, PCA+SDA, and PCA+SVM. • Even though the input to LDA, SDA and SVM are dimensionality reduced principal components, these three algorithms are able to find good decision boundaries with large number of PCA components whereas PCA is providing very low accuracy for the same....
[...]
...Contrary to existing literature, the results show that PCA provides comparable but slightly better performance compared to PCA+LDA, PCA+SDA and PCA+SVM....
[...]
...Moreover, when a range of training databases and principal components is available, PCA provides the best generalization performance compared to LDA, SDA, and SVM....
[...]
...With 10 PCA features, PCA+SVM is not able to learn the classifier, however, for 20-100 PCA features, the performance is increasing and is in the range of 78- 86%....
[...]
...The results showed that for single database testing, SVM [17] and Adaboost yield best results followed by linear discriminant analysis approaches whereas for cross database tests, PCA+LDA provided the best results and SVM was the lowest....
[...]

Journal Article•DOI•

The FERET evaluation methodology for face-recognition algorithms

[...]

P.J. Phillips, Hyeonjoon Moon, Syed A. Rizvi¹, Patrick J. Rauss²•Institutions (2)

College of Staten Island¹, United States Army Research Laboratory²

01 Oct 2000-IEEE Transactions on Pattern Analysis and Machine Intelligence

TL;DR: Two of the most critical requirements in support of producing reliable face-recognition systems are a large database of facial images and a testing procedure to evaluate systems.

...read moreread less

Abstract: Two of the most critical requirements in support of producing reliable face-recognition systems are a large database of facial images and a testing procedure to evaluate systems. The Face Recognition Technology (FERET) program has addressed both issues through the FERET database of facial images and the establishment of the FERET tests. To date, 14,126 images from 1,199 individuals are included in the FERET database, which is divided into development and sequestered portions of the database. In September 1996, the FERET program administered the third in a series of FERET face-recognition tests. The primary objectives of the third test were to 1) assess the state of the art, 2) identify future areas of research, and 3) measure algorithm performance.

...read moreread less

4,816 citations

"Is gender classification across eth..." refers methods in this paper

...The experiments were performed on around 1800 mugshot images of the FERET database [12] and reported 3....
[...]
...The experiments were performed on around 1800 mugshot images of the FERET database [12] and reported 3.6% error with 80% training and 20% testing....
[...]
...On frontal images from the FERET [12] database and 80%-20% non-overlapping train-test partitioning, the algorithm showed the maximum accuracy of 93%....
[...]
...It contains images from the CMU PIE [14], Georgia Tech [1], GTAV [16] and FERET [12] face databases with neutral expression, minimum illumination variation, and no occlusion....
[...]

The AR face database

[...]

Aleix M. Martinez

01 Jan 1998

3,650 citations

Additional excerpts

...of female face images face images AR [7] 527 453 Indian Face 250 250 FRGC [11] 600 407 Combined [15] 962 576 Notre Dame [4] 1712 580 Plastic Surgery [15] 195 1600 Total (8112) 4246 3866...
[...]

Journal Article•DOI•

PCA versus LDA

[...]

Aleix M. Martinez¹, Avinash C. Kak¹•Institutions (1)

Purdue University¹

01 Feb 2001-IEEE Transactions on Pattern Analysis and Machine Intelligence

TL;DR: In this article, the authors show that when the training data set is small, PCA can outperform LDA and, also, that PCA is less sensitive to different training data sets.

...read moreread less

Abstract: In the context of the appearance-based paradigm for object recognition, it is generally believed that algorithms based on LDA (linear discriminant analysis) are superior to those based on PCA (principal components analysis). In this communication, we show that this is not always the case. We present our case first by using intuitively plausible arguments and, then, by showing actual results on a face database. Our overall conclusion is that when the training data set is small, PCA can outperform LDA and, also, that PCA is less sensitive to different training data sets.

...read moreread less

3,102 citations