Deepfake detection

Project Title

Deepfake detection

Project Area of Specialization Artificial IntelligenceProject Summary

DeepFake is a generative deep learning algorithm that creates or changes facial features in a very realistic way, making it hard to tell the real from the fake features. It can be used to make movies look better, as well as to spread false information by imitating famous people.There are many different ways to make a deep fake and In this project, we look at all of them and put them into separate categories. Using DeepFake datasets, models are trained and then tested with experiments to see if they are useful. DeepFakes are a type of facial manipulation that allows people to change their whole faces, their identities, their attributes, and their expressions. We also talk about the trends in the available DeepFake datasets, with a focus on how they've changed.

Project Objectives

This project is based on the survay, after collecting data from different survays and researchs we will develop deepfake detector with possible high accuraccy. The main objective of this project is to detect fake images with handsome accuracy.

Project Implementation Method

By using deep learning we will train CNN model with experimental parameters to achieve high accuracy using public datasets such as FF++ , DFDC.

It will be implemented on Python with different libs such as Tensorflow , Numpy and pandas etc.

Benefits of the Project

Law enforcment and other forensics agencies can use this model for the detection of the fake images. 

By implementing our model into application laymen can check on his mobile that video that he is seeing is fake or real.

False video can be created for the criminal purposes , by using this tool we can overcome this problem.

Technical Details of Final Deliverable

Deep learning will be used to make a general DeepFake detection model. There will also be a discussion of the problems with making and detecting DeepFakes. As a result of this survey based projectI, we hope that we'll be able to speed up the development of new DeepFake-based imaging tools in the future.

Final Deliverable of the Project Software SystemCore Industry ITOther Industries Security Core Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development GoalsRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	49000
Colab Pro+	Equipment	5	9800	49000