AUTOMATIC AGE AND GENDER

ESTIMATION USING FACIAL DATA

PROJECT GUIDE
MRS.J.SUDHA RANI

TEAM MEMBERS
P. GIRISH KUMAR (TEAM LEADER) 16H61A0458
K. KALYAN 17H65A0404
S.BHARADWAJ 16H61A0445
M.SHASHI KUMAR 15H65A0408
 AGENDA
 Abstract

 Introduction

 Literature Survey

 Problem Statement

 Objectives

 Proposed Methodology

 Block Diagram

 DNN

 References
Abstract

 Age estimation and gender classification using facial data is very

important for the identification of human age.
 Now a day’s face is being detected at the time of dead bodies, child
marriage cases, at the hospitals, government authorities.
 Age and gender estimation can be made by the identification of
eyes, nose, lips, ears and hair color.
 Sometimes we cannot detect the age because many people will be
applying facials, make up and color to their hair.
 To overcome this, we need to add some of the features for the age
estimation. Mostly for the classification purpose we are using neural
network algorithm.
 For the feature identification we are using feature extraction
process.
 INTRODUCTION
 Age and gender, two of the key facial attributes, play a very
foundational role in social interactions, making age and gender
estimation from a single face image an important task in intelligent
applications, such as access control, human-computer interaction, law
enforcement, marketing intelligence and visual surveillance, etc.
 Automatic age and gender classification has become relevant to an
increasing amount of applications, particularly since the rise of social
platforms and social media.
Deep Learning
 Deep learning is the hottest field in AI right now.
 From Google Duplex assistant to Tesla self- driving cars the
applications are endless.
 In the past 10 years, machine learning and artificial intelligence have
shown tremendous progress.
 Recently deep learning plays major role for
1.Explosion of huge data for age and face
2.Cheap Computing cost
3.Improvement in ML algorithms for face type projects
Recently we came across Quividi which is an AI software
application which is used to detect age and gender of users who
passes by based on online face analyses and automatically starts
playing advertisements based on the targeted audiences age and
gender.
 Literature Survey

 Early methods for age estimation are based on calculating ratios

between different measurements of facial features.
 Once facial features (e.g. eyes, nose, mouth, chin, etc.) are
localized and their sizes and distances measured, ratios between
them are calculated and used for classifying the face into different
age categories according to hand-crafted rules.
 More recently, uses a similar approach to model age progression in
subjects under 18 years old.
 As those methods require accurate localization of facial features, a
challenging problem by itself, they are unsuitable for in-the-wild
images which one may expect to find on social platforms.
Problem Statement

 To Design & Implement a technique for Automatic detection

of age and gender of a person using MATLAB software.
Objectives

 To study different existing techniques and their drawbacks for

age and gender detection.
 To study DNN(Deep Neural Network) based Classification
algorithm.
 To implement a MATLAB code for classification a person to
get gender and age.
Proposed Method

Input Images Pre- Feature

processing Extraction

Datasets Training Process NN Algorithm

Fig : Block Diagram

 System contains 3 major task-

 Taking an input image , processing on it , and an output in terms
of audio for age and gender.
Features
 Basic features of any image is color , texture and shape.
 Depending on our requirement we will select textural features.
 Deep learning also helps to extract the features using autoencoder.
DNN Classifier
Input Test Image

Feature Extraction from face

Training Data

Auto Encoders
Features From Trained
Data
Soft Max Layer

Deep Neural Network

Post-Processing(Audio)
Conclusion
 Though many previous methods have addressed the
problems of age and gender classification.
 But with the help of proposed work we got the results near
to optimal performance which can be shown by both
subjective and objective analysis.
 Deep learning is one of the best technique we used is
shown here.
ADVANTAGES
1. Time complexity is less
2. Higher accuracy than existing techniques
3. Robust : natural variations in data is automatically learned.
4. Generalizable : Same DNN can be used for different
applications.
5. Scalability: If we increase the data then also we will get the
same algorithm will work on that data without any further
modifications in algorithm.
MATLAB toolboxes Used

 Deep Learning Toolboxes and image processing toolboxes

 Convolution Neural Networks toolbox
 Autoencoders for deep learning toolboxes.
References
[1] T. Ahonen, A. Hadid, and M. Pietikainen. Face description with local binary patterns:
Application to face recognition. Trans. Pattern Anal. Mach. Intell., 28(12):2037–2041, 2006.
[2] S. Baluja and H. A. Rowley. Boosting sex identification performance. Int. J. Comput. Vision,
71(1):111–119, 2007.
[3] A. Bar-Hillel, T. Hertz, N. Shental, and D. Weinshall. Learning distance functions using
equivalence relations. In Int. Conf. Mach. Learning, volume 3, pages 11–18, 2003.
[4] W.-L. Chao, J.-Z. Liu, and J.-J. Ding. Facial age estimation based on label-sensitive learning
and age-oriented regression. Pattern Recognition, 46(3):628–641, 2013.
[5] K. Chatfield, K. Simonyan, A. Vedaldi, and A. Zisserman. Return of the devil in the details:
Delving deep into convolutional nets. arXiv preprint arXiv:1405.3531, 2014. 3