Video Based Evidence Analysis and Extraction in Digital Forensic Investigation

Video Based Evidence Analysis and Extraction in Digital Forensic Investigation

Publication Date : 2024-05-02
Author(s) :

S. Venkata Kiran, B. Himabindhu, G. Revathi, K. Sudarsan Reddy, K. Mohan Krishna, K. Rosi Reddy
Conference Name :

Sri Venkatesa Perumal College of Engineering and Technology
Abstract :

With the rapid advancement of technology and the ease of creating fake content, the manipulation of media has become widespread in recent times. The emergence of AI-altered videos, known as Deepfakes, poses a significant threat to media integrity as they are increasingly being produced and disseminated across various social media platforms. Detecting such Deepfakes has become a major challenge in digital forensics. In this paper, we propose an approach for detecting Deepfake videos using Recurrent Neural Network (RNN) algorithms. The proposed method and its steps are elaborated upon in detail. We achieved an accuracy of 91% for the developed Deep Learning (DL) model using the Celeb-DF dataset. Finally, we present the results of fake or real detection in the provided videos, demonstrating the effectiveness of our approach.

No. of Downloads :

5