International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 05 | May 2022
p-ISSN: 2395-0072
www.irjet.net
Gender and age classification using deep learning Dr. Vandana S Bhat1, Shivani A Patilkulkarni2, Pratiksha Karkannavar3, Aditi M4, Madhura Chandunavar5 1Dr.
Vandana S Bhat, Professor, Department of Information Science and Engineering, SDMCET, Dharwad, Karnataka 2,3,4,5 Students, Department of Information Science and Engineering, SDMCET, Dharwad, Karnataka ---------------------------------------------------------------------***--------------------------------------------------------------------2. LITERATURE SURVEY Abstract - 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. We will be using convolutional neural networks as a deep learning technique to predict age and gender of the facial image. The benchmark dataset that we will be using to train the model is the UTK Face dataset which is obtained from Kaggle. It is not a pre-processed dataset. At the end, an offline mobile application will be built to predict age and gender of the given input image i.e., facial image.
The survey of different papers related to Age and Gender classification has been carried out in this section. [1] Malik, Rathi et al. (IRJET,2020) proposed a model for age estimation and gender recognition using Convolutional Neural Network as a deep learning model. They have used the Kaggle dataset which is the greatest available dataset of human faces for training purposes. They have used Keras high level API of TensorFlow. Keras is used for building and training of their proposed model. They have used the VGG face model for age and gender estimation. They have used VGG-16 architecture that runs on cropped images of faces.
Key Words: Age, Gender, Human-Computer Interaction, Android Application, Android Studio, Deep Learning, Convolutional Neural Network, Facial Image.
[2] Agarwal, Dixit (Springer, 2020) proposed research to estimate a person’s gender and age on the basis of visualization. They explain that to estimate age and gender from a facial image is a difficult task due to variations, lighting and other conditions in the face image. They proposed a methodology which uses Convolutional Neural Network as model and applies Principal Component Analysis technique to reduce the extracted features dimension. This work is done on the IMDB-WIKI dataset as well as their own dataset.
1. INTRODUCTION Age and gender are two key facial attributes which play a very important role in social collaborations, making age and gender estimation from a face picture a vital task in smart applications. In this growing industry, age and gender recognition has become one of the important models for computer vision applications for example human global interaction and passive demographic data collection. Recently the interest in the advertising industry for launching specific demographic specific marketing and targeted advertising and public pages, has attracted the attention of more researchers in the field of computer vision to the field of age and gender classification. Human face may be a storehouse of various information about personal characteristics, including identity, colour, expression, gender, age, etc. We will be developing an offline mobile application which classifies age and gender according to the input image given. Entire paper is organized into six sections. Section I has the introduction for Age and Gender classification. Section II includes the literature survey of various Age and Gender detection papers. Section III contains methodology of the proposed project. Section IV explains experimental results of the Age and Gender classification. Section V elaborates the conclusion of the proposed project followed by Section VI which contains acknowledgement.
© 2022, IRJET
|
Impact Factor value: 7.529
[3] Saxena, Singh et al. (IEEE,2021) proposed research paper that systematically describes the process, the different methods and algorithms that can be used, the most accurate method. They will also highlight its importance and how it can be useful in our daily lives. The main focus of this paper is to create a gender and age identifier that can predict the gender and age of a person's face in an image using in-depth reading of the targeted database. They get the most effective predictions and results by overcoming the problem of accuracy and timing. [4] Chao Yin et.al (IEEE, 2019) proposed paper that the Conditional Probability Neural Network (CPNN) is a distributed learning algorithm that is used to measure age using facial image. It follows a three-layer neural network system with targets and a vector of conditional features. The training method of this program uses the relationship between image of the face and distribution of its label via the neural network. The method that was previously used assumed that measurements should be used according to the
|
ISO 9001:2008 Certified Journal
|
Page 3395