Technical Journal, Vol. 16 No. 4, 2022.
Original scientific paper
https://doi.org/10.31803/tg-20220403080215
AFMB-Net: DeepFake Detection Network Using Heart Rate Analysis
A. Vinay
; PES University, Ring Road Campus, Bangalore, Karnataka, India
Nipun Bhat
; PES University, Ring Road Campus, Bangalore, Karnataka, India
Paras S. Khurana
orcid.org/0000-0003-4701-0217
; PES University, 100 Feet Ring Road, BSK III Stage, Ring Road Campus, Bangalore-560085, Karnataka, India
Vishruth Lakshminarayanan
; PES University, Ring Road Campus, Bangalore, Karnataka, India
Vivek Nagesh
; PES University, Ring Road Campus, Bangalore, Karnataka, India
S. Natarajan
; PES University, Ring Road Campus, Bangalore, Karnataka, India
T. B. Sudarshan
; PES University, Ring Road Campus, Bangalore, Karnataka, India
Abstract
With advances in deepfake generating technology, it is getting increasingly difficult to detect deepfakes. Deepfakes can be used for many malpractices such as blackmail, politics, social media, etc. These can lead to widespread misinformation and can be harmful to an individual or an institution’s reputation. It has become important to be able to identify deepfakes effectively, while there exist many machine learning techniques to identify them, these methods are not able to cope up with the rapidly improving GAN technology which is used to generate deepfakes. Our project aims to identify deepfakes successfully using machine learning along with Heart Rate Analysis. The heart rate identified by our model is unique to each individual and cannot be spoofed or imitated by a GAN and is thus susceptible to improving GAN technology. To solve the deepfake detection problem we employ various machine learning models along with heart rate analysis to detect deepfakes.
Keywords
Convolutional Neural Network (CNN); Celeb-DF; Deepfake; Frame Normalisation; Generative Adversarial Networks (GAN); Motion Magnification Spatio-Temporal Map (MMST Map); MBConv Blocks; PhotoPlethysmoGraphy (PPG)
Hrčak ID:
283786
URI
Publication date:
23.9.2022.
Visits: 1.524 *