Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

DeepFake detection

using deep learning

TEAM MEMBERS
Noopur Sharma
Vartika Singh
Vivashwat Thakur
TABLE OF CONTENTS

01 02 03 04 05
Technolog- DeepFake Conclusion
Introduction How DeepFake
y used detection
is Created ?
&dataset
Introduction
What are Deepfakes?

Deepfake is a technique that uses deep learning algorithms to create fak


media usually by swapping a person’s face/audio from a source into anot
person’s face in a target, with a resulng a fake output that is somemes ha
to detect.In our project, we aim to detect such media with the use of ima
processing and deep learning methods.
We have developed a DeepFake Detection system using advanced deep-
learning techniques, including ResNext and LSTM. The integration of the
trained model with the Frontend UI is achieved through ReactJs, while Flask
serves as the Backend. It's purpose is to predict the authenticity of videos and
photos discerning between FAKE and REAL, providing users with a confidence
ratio.
How DeepFake is created?
Deepfakes are created using deep learning methods in which they aim to replac
face of a targeted person by the face of someone else in an image or video.

Deepfake videos are often made using avariational auto-encoder (VAE)


and a facial recognition algorithm.
EXAMPLE
Technology used
Python,HTML,CSS
JupyterNotebook
Tensorflow

Frontend UI-ReactJs
Backend-Flask
DeepFake Datasets Used
DeepFake-TIMIT

FaceForensics++

Celeb-DF
DeepFake
Detection
Deepfake Detection To successfully build a robust
neural network capable of detecting complex
deepfakes a large dataset of training images is needed.

After training the CNN with the dataset, the CNN is able
to detect deepfake images with over 80 % confidence
rate
What is CNN?
A Convolutional Neural Network (CNN) is a type of deep learning algorithm that is particularly well-suited for image
recognition and processing tasks. It is made up of multiple layers, including convolutional layers, pooling layers, and
connected layers.
Long short-term memory LSTM is an artificial
recurrent neural network RNN architecture used
in the field of deep learning. Unlike standard
feedforward neural networks, LSTM has feedback
connections. It can not only process single data points
such as images, but also entire sequences of data
such as speech or video.
CONCLUSION
It seem we’re only on the cusp
of what deep fake technology
could lead to our future in the
coming decade.While Deep fakes
on their own seems harmless,
when put in the wrong hands it
can cause serious controversy.
THANK YOU

You might also like