scholarly journals Does Face Recognition Error Echo Gender Classification Error?

2021 ◽  
Author(s):  
Ying Qiu ◽  
Vitor Albiero ◽  
Michael C. King ◽  
Kevin W. Bowyer
Keyword(s):  
Face Recognition ◽  
Classification Error ◽  
Gender Classification ◽  
Recognition Error

scholarly journals An IOT Method for Reducing Classification Error in Face Recognition With the Commuted Concept of Conventional Algorithm

2019 ◽  
Vol 8 (11) ◽  
pp. 1-7
Keyword(s):  
Face Recognition ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Conventional Approach ◽  
Combined Effect ◽  
Classification Error ◽  
Naive Bayes Classifier ◽  
Bayes Classifier ◽  
Naïve Bayes Classifier ◽  
Conventional Algorithm

The initial work has discussed the conventional approach of algorithms along with their drawbacks and features. Apart from that types of face recognition methodologies have been discussed with application of IOT trends. We specifically depicts a descriptive idea about working and applications of all conventional algorithms which have been commuted concept wise in proposed methodologies section of our work Our work consists of literature survey so we can provide a reason for the previous work and get basic ground for performing and implementing proposed work. One of a common procedures of face detection has been discussed that’s has been worked out in past with accuracy .The observation in this work leads to propose the method by commuting the conventional algorithm, Basically the work done with conventional approach has been discussed in this section with a strong focus over the role of Iot in face recognition and what is importance of Iot in this domain and what changes Iot concept has bring about as far as face recognition with different approach has been concerned . Not only PCA concept has been commuted but along with Pca, Svm, naïve bayes classifier, DCT, Gabor, neural network efficiency and their combined effect has been performed and analyzed later. Our work has been focusing around commuted concept of conventional algorithms so this particular chapter is very much important to discuss the conventional methodologies perform by classical mathematically implemented techniques for classifications. With the help of the analysis we will discuss the problem formulation and comparison of proposed work with existing work .So our work is basically about the problem existing with conventional algorithm for classifications and what lead us to propose the commuted concept further to deal or minimize the effect of that particular problem ,Our work is not primarily based on face recognition but to calculate the classification error through conventional algorithm and then compare it with our proposed commuted concept and combined effect of conventional algorithms as well, like PCA+SVM PCA+ Kernel SVM, Commuted Concept of PCA +Naïve bayes Classifier .We have gone through with different cases to ensure the minimization of classification error through proposed method .The goal of the work is to associate the application of

Manipulating Face Gender

1998 ◽  
Vol 06 (03) ◽  
pp. 219-239 ◽  
Author(s):  
Kenneth A. Deffenbacher ◽  
Cheryl Hendrickson ◽  
Alice J. O'Toole ◽  
David P. Huff ◽  
Hervé Abdi
Keyword(s):  
Face Recognition ◽  
Computational Model ◽  
Face Image ◽  
Gender Classification ◽  
Statistical Features ◽  
Female Adult ◽  
Recognition Experiment ◽  
Accuracy Data ◽  
Speed And Accuracy ◽  
Weighted Combination

Previous research has shown that faces coded as pixel-based images may be constructed from an appropriately weighted combination of statistical "features" (eigenvectors) which are useful for discriminating members of a learned set of images. We have shown previously that two of the most heavily weighted features are important in predicting face gender. Using a simple computational model, we adjusted weightings of these features in more masculine and more feminine directions for both male and female adult Caucasian faces. In Experiment 1, cross-gender face image alterations (e.g., feminizing male faces) reduced both gender classification speed and accuracy for young adult Caucasian observers, whereas same-gender alterations (e.g., masculinizing male faces) had no effect as compared to unaltered controls. Effects on femininity-masculinity ratings mirrored those obtained on gender classification speed and accuracy. We controlled statistically for possible effects of image distortion incurred by our gender manipulations. In Experiment 2 we replicated the same pattern of accuracy data. Combined, these data indicate the psychological relevance of the features derived from the computational model. Despite having different effects on the ease of gender classification, neither sort of gender alteration negatively impacted face recognition (Experiment 3), yielding evidence for a model of face recognition wherein gender and familiarity processing proceed in parallel.

MCE-BASED FACE RECOGNITION

2001 ◽  
Vol 15 (08) ◽  
pp. 1311-1327 ◽  
Author(s):  
LIANG-HUA CHEN ◽  
SHAO-HUA DENG ◽  
HONG-YUAN LIAO
Keyword(s):  
Neural Network ◽  
Face Recognition ◽  
Face Detection ◽  
Detection Algorithm ◽  
Recognition Algorithm ◽  
Classification Error ◽  
Complex Scenes ◽  
Minimum Classification Error ◽  
Complete Procedure ◽  
Human Faces

This paper proposes a complete procedure for the extraction and recognition of human faces in complex scenes. The morphology-based face detection algorithm can locate multiple faces oriented in any direction. The recognition algorithm is based on the minimum classification error (MCE) criterion. In our work, the minimum classification error formulation is incorporated into a multilayer perceptron neural network. Experimental results show that our system is robust to noisy images and complex background.

scholarly journals Face Detection and Gender Classification using Facial Features

2019 ◽  
Vol 8 (4) ◽  
pp. 6670-6674
Keyword(s):  
Face Recognition ◽  
Face Detection ◽  
Recognition System ◽  
Facial Features ◽  
Gender Classification ◽  
Biometric System ◽  
Detection Techniques ◽  
The Face ◽  
And Gender ◽  
Two Stages

Face Recognition is the most important part to identifying people in biometric system. It is the most usable biometric system. This paper focuses on human face recognition by calculating the facial features present in the image and recognizing the person using features. In every face recognition system follows the preprocessing, face detection techniques. In this paper mainly focused on Face detection and gender classification. They are performed in two stages, the first stage is face detection using an enhanced viola jones algorithm and the next stage is gender classification. Input to the video or surveillance that video converted into frames. Select few best frames from the video for detecting the face, before the particular image preprocessed using PSNR. After preprocessing face detection performed, and gender classification comparative analysis done by using a neural network classifier and LBP based classifier

scholarly journals Support Vector Machine Based Feature Extraction For Gender Recognition From Objects Using Lasso Classifier

2020 ◽  
Author(s):  
Damodara Krishna Kishore Galla ◽  
BabuReddy Mukamalla ◽  
Rama Prakasha Reddy Chegireddy
Keyword(s):  
Support Vector Machine ◽  
Feature Selection ◽  
Face Recognition ◽  
Selection Process ◽  
Recognition System ◽  
Classification Model ◽  
Support Vector ◽  
Gender Classification ◽  
Accuracy Score ◽  
Lasso Regression

Abstract The blind people has their difficulty to identify the object moving around them, therefore with a high accuracy score object detection and human face recognition system will helps them in identifying the things around them with ease. Facial record images are immobile an difficult assignment for biometric authentication systems due to various types of characteristics are dimensions, pose, expressions, illustrations and age etc. In facial and other united images includes different objects classifications. In this research article, a minimum distance trainer for feature selection by accessing SVM feature optimization process. For feature selection process SVM (support vector machine) was considered for improving its feature interpretability and computational efficiency., then LASSO classifier applied to perform object recognition and gender classification. Original face image database used for the gender classification. This approach was implemented with dual classification model (1) Recognizing or classifying human faces from various objects and (2) Classifying gender through face recognition] is made possible with the help of combining modified SIFT feature in combination with ridge regression (RR), elastic net (EN), lasso regression(LR) and lasso regression with Gaussian Support Vector Machines (LRGS) based classification.

scholarly journals Support Vector Machine Based Feature Extraction for Gender Recognition From Objects Using Lasso Classifier

2020 ◽  
Author(s):  
Damodara Krishna Kishore Galla ◽  
BabuReddy Mukamalla ◽  
Rama Prakasha Reddy Chegireddy
Keyword(s):  
Face Recognition ◽  
Recognition System ◽  
Classification Model ◽  
Support Vector ◽  
Gender Classification ◽  
Accuracy Score ◽  
Lasso Regression ◽  
Vehicle Recognition ◽  
And Gender ◽  
Human Faces

Abstract The blind people has their difficulty to identify the object moving around them, therefore with a high accuracy score object detection and human face recognition system will helps them in identifying the things around them with ease. In this research article,a minimum distance trainer for feature selection by accessing SVM feature optimization process, then LASSO classifier applied to perform object recognition and gender classification. Database of 100 images (50 male and 50 female face images considered from 5 different databases) and 10 categories of vehicle types are used for gender and vehicle recognition and classification. Original face image database used for the gender classification. This approach was implemented with dual classification model [(1) Recognizing or classifying human faces from various objects and (2) Classifying gender through face recognition] is made possible with the help of combining modified SIFT feature in combination with ridge regression (RR), elastic net (EN), lasso regression(LR) and lasso regression with Gaussian Support Vector Machines (LRGS) based classificatioins. The final classification results accurate are as follows RR- 89.6%, EN- 93.5%, LR-93.2% and the proposed approach is LRGS with 98.4% accurate detection rate with rediction names.

Sign in / Sign up

Export Citation Format

Share Document