Deep learning based face attributes recognition
Loading...
Date
2018-06
Authors
Mohamad Hazim Saidi
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Face Recognition is a recently developing technology with numerous real life
applications. The goal of this Final Year Project is to create a complete Face Attributes
Recognition for security or facility. The automated face identification application is helpful in
assisting forensic to survey an area with the implementation of Machine Learning (ML). It was
once a difficult challenge due to uncertainties in the captured such as high variation of pose and
obstruction corresponding to voluntary and involuntary factors. With the introduction of Deep
Learning (DL), the concept of Convolutional Neural Network (CNN) that was once an idea can
be realized. In this project, Fast CNN architecture is used as the core engine to power the face
attributes recognition that aims to help users to identify its. The users can identify the face
attributes including gender, glasses and facial hair. The face attributes recognition also can be
performed well in identifying properly and improperly frontalized faces. Optimization is
performed by experimenting in stages with several training parameters to obtain the best value
for this unique purpose. Using this architecture, the best performance of training algorithm can
be produced in order to recognize face attributes. Combined-algorithm based optimizers plays
an important role in optimizing the training algorithm. The addition of convolutional layer is
also essential in order to extract related facial features of facial images.