Analysis and Detection of Deep Fake Images using Deep Learning Techniques

Abstract: The technological advancements in the fields of machine learning and deep learning gave a way for realistic AI generated fake images called as Deep Fakes. They refer to manipulated images, videos, text and audio that cannot be distinguished from authentic ones. So, we can understand that there are enough potentials and also respective apprehensions from these deep fakes. Now a days, as we can observe more malafide utilizations of these deep fakes, which are becoming the threat to national security, individual privacy and democracy. So, the necessities for the deep fake detection techniques have become need of the hour. In recent days, we have plethora of platforms to generate deep fakes but we are lacking in the holistic deep fake detection techniques. To bridge this deficit, a lot of researches have been conducted so far using different ways like blinking of eyes, inconsistent head poses, background compressions, mouth movements, face forgery techniques, illumination and resolution effects etc. with the help of machine learning and deep learning techniques such as support vector machines (SVM), convolution neural networks(CNN) and CNN with DenseNet. As these deepfakes are posing many challenges, there is a need to sort out the best way as soon as possible by increasing the accuracy of the model which could help the humanity in constructive utilization of the technology.

Outline of the Tutorial:

  • Introduction to Machine Learning and Deep Learning
  • Introduction to Deepfakes
  • Typology of deepfakes and their applications
  • Types of GAN’s
  • Comparative Analysis of GAN’s
  • Generative Techniques
  • Detecting of Deepfakes
  • Application Areas of Deepfakes in real time environment
  • Issues with Deepfakes

Expected Length of the Tutorial: Half day (2-3 hours)

Level of Tutorial: Introductory (Discussion on few case studies and implementation of deep fakes using python modules)


Mr. K. RaviKanth, Assistant Professor and R&D Coordinator – CSE; IIIT- RGUKT Basar
Email ID: ,
Mobile: +91-9247448766