ENHANCING VIDEO FORGERY DETECTION WITH DEEP CONVOLUTIONAL NEURAL NETWORKS

Abstract

A lot of people are worried about video forgeries and how they could affect digital forensics, media integrity, and security because of how sophisticated editing tools are getting. Because subtle changes are so hard to spot using conventional detection methods, it is a demanding undertaking. The primary goal of this paper is to examine how Deep Convolutional Neural Networks (DCNN) can improve video counterfeit detection. With the help of deep learning techniques, the proposed model can detect cases of deepfake changes, splicing, and frame tampering with a remarkable degree of accuracy. Results from the experiments indicate that deep convolutional neural networks (DCNN) outperform more traditional approaches, which could make them valuable in forensic investigations.