Fake Face Detection by Deep Learning Technique Test phase, srs, design phase and source code final deliverable
Project Domain / Category
AI/Deep Learning/Machine Learning
Abstract / Introduction
With the growing era of social media, it is difficult to pick out the actual from fake whether it’s miles any news or face/video of any celeb, baby-kisser and so forth. Also, the faux or manipulated faces and motion pictures are being generated exceedingly which can be harder to stumble on by means of traditional manner of software program or methods. Therefore, Deep Learning that’s a subset of Machine Learning can be hired to pick out the actual or fake pictures/faces/movies efficiently. We are going to hit upon the faces as actual or fake from the given dataset. As there are already multiple photos with faces which can be generated via various softwares inside the dataset. In this project, we will use Convolutional Neural Network (CNN) to detect the fake faces from an online database (https://www.Kaggle.Com/xhlulu/140k-actual-and-faux-faces), we also can talk approximately the various datasets and their utilization. The photographs are fed into the detector, and it’s going to then locate whether or not the picture this is fed into it’s miles actual or faux.
This is indeed a completely interesting mission however calls for intensive study of deep studying, neural networks. The following hyperlinks may additionally assist you in higher understanding:
Convolutional Neural Network:
https://towardsdatascience.Com/convolutional-neural-networks-defined-9cc5188c4939
Deep Learning for Face Recognition:
https://machinelearningmastery.Com/advent-to-deep-learning-for-face-popularity/
Dataset:
https://www.Kaggle.Com/xhlulu/140k-actual-and-faux-faces
Functional Requirements:
The following are the purposeful necessities of the assignment:
1. The tool based totally utility/software program have to down load the given Dataset that includes the database of real and fake snap shots.
2. The system should be any CNN version that carries hidden layers for the fake face detection.
3. Whenever, any face is given as an input into the detection system, it identifies as real or fake as output.
Four. The device must be capable of stumble on the fake faces generated with the aid of any of the android apps like FaceApp, FaceSwap, Wombo etc.
Tools:
● Python (programming language)
● Keras (API)
● Tensorflow (open supply software library for machine studying)
● Jupyter Notebook (open source net application)
● Matplotlib (library)
● Numpy (library for the python)