FakeBuster: A deepfakes detection tool for video conferencing scenarios

Vineet Mehta, Parul Gupta, Ramanathan Subramanian, Abhinav Dhall

Research output: A Conference proceeding or a Chapter in BookOther chapter contributionpeer-review

Abstract

This paper proposes FakeBuster, a novel DeepFake detector for (a) detecting impostors during video conferencing, and (b) manipulated faces on social media. FakeBuster is a standalone deep learning- based solution, which enables a user to detect if another person's video is manipulated or spoofed during a video conference-based meeting. This tool is independent of video conferencing solutions and has been tested with Zoom and Skype applications. It employs a 3D convolutional neural network for predicting video fakeness. The network is trained on a combination of datasets such as Deeperforensics, DFDC, VoxCeleb, and deepfake videos created using locally captured images (specific to video conferencing scenarios). Diversity in the training data makes FakeBuster robust to multiple environments and facial manipulations, thereby making it generalizable and ecologically valid.

Original languageEnglish
Title of host publication26th International Conference on Intelligent User Interfaces, IUI 2021 Companion
EditorsKatrien Verbert, Dennis Parra
Place of PublicationUnited States
PublisherAssociation for Computing Machinery (ACM)
Pages61-63
Number of pages3
ISBN (Electronic)9781450380188
DOIs
Publication statusPublished - 14 Apr 2021
Externally publishedYes
Event26th International Conference on Intelligent User Interfaces: Where HCI Meets AI, IUI 2021 - Virtual, Online, United States
Duration: 14 Apr 202117 Apr 2021

Publication series

NameInternational Conference on Intelligent User Interfaces, Proceedings IUI

Conference

Conference26th International Conference on Intelligent User Interfaces: Where HCI Meets AI, IUI 2021
CountryUnited States
CityVirtual, Online
Period14/04/2117/04/21

Fingerprint Dive into the research topics of 'FakeBuster: A deepfakes detection tool for video conferencing scenarios'. Together they form a unique fingerprint.

Cite this