Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems

Alakananda Mitra; Saraju P. Mohanty; Peter Corcoran; Elias Kougianos

doi:10.1109/OCIT53463.2021.00039

Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems

Alakananda Mitra
, Saraju P. Mohanty
, Peter Corcoran
, Elias Kougianos

Electrical & Electronic Engineering

University of North Texas

Research output: Chapter in Book or Conference Publication/Proceeding › Conference Publication › peer-review

13 Citations (Scopus)

Abstract

Face Morphing has emerged as a pervasive attack of Facial Recognition Systems. The rapid growth of Generative Adversarial Networks takes it to a complete new level. Deepfake or deep neural network based face morphing, a.k.a deep-morph attack, presents a significant threat to Facial Recognition System. In this paper, we propose a novel Convolutional Neural Network based detection method of deep morphed deepfake images which is suitable for IoT environments in smart cities. A high accuracy of 94.83% has been achieved for the DeepfakeTIMIT HQ dataset. This lightweight and fast network is a natural choice for IoT environments.

Original language	English
Title of host publication	Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	149-154
Number of pages	6
ISBN (Electronic)	9781665416641
DOIs	https://doi.org/10.1109/OCIT53463.2021.00039
Publication status	Published - 2021
Event	19th Orissa Information Technology Society International Conference on Information Technology, OCIT 2021 - Bhubaneswar, India Duration: 16 Dec 2021 → 18 Dec 2021

Publication series

Name	Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021

Conference

Conference	19th Orissa Information Technology Society International Conference on Information Technology, OCIT 2021
Country/Territory	India
City	Bhubaneswar
Period	16/12/21 → 18/12/21

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 11 Sustainable Cities and Communities

Keywords

Convolutional Neural Network
Deep Learning
Deep-fake
Deep-Morph
Facial recognition System
Smart City

Access to Document

10.1109/OCIT53463.2021.00039

Cite this

Mitra, A., Mohanty, S. P., Corcoran, P., & Kougianos, E. (2021). Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems. In Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021 (pp. 149-154). (Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/OCIT53463.2021.00039

Mitra, Alakananda ; Mohanty, Saraju P. ; Corcoran, Peter et al. / Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems. Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 149-154 (Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021).

@inproceedings{59f8dba87da748c4a06fb534fcb92105,

title = "Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems",

abstract = "Face Morphing has emerged as a pervasive attack of Facial Recognition Systems. The rapid growth of Generative Adversarial Networks takes it to a complete new level. Deepfake or deep neural network based face morphing, a.k.a deep-morph attack, presents a significant threat to Facial Recognition System. In this paper, we propose a novel Convolutional Neural Network based detection method of deep morphed deepfake images which is suitable for IoT environments in smart cities. A high accuracy of 94.83\% has been achieved for the DeepfakeTIMIT HQ dataset. This lightweight and fast network is a natural choice for IoT environments.",

keywords = "Convolutional Neural Network, Deep Learning, Deep-fake, Deep-Morph, Facial recognition System, Smart City",

author = "Alakananda Mitra and Mohanty, \{Saraju P.\} and Peter Corcoran and Elias Kougianos",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 19th Orissa Information Technology Society International Conference on Information Technology, OCIT 2021 ; Conference date: 16-12-2021 Through 18-12-2021",

year = "2021",

doi = "10.1109/OCIT53463.2021.00039",

language = "English",

series = "Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "149--154",

booktitle = "Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021",

}

Mitra, A, Mohanty, SP, Corcoran, P & Kougianos, E 2021, Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems. in Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021. Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021, Institute of Electrical and Electronics Engineers Inc., pp. 149-154, 19th Orissa Information Technology Society International Conference on Information Technology, OCIT 2021, Bhubaneswar, India, 16/12/21. https://doi.org/10.1109/OCIT53463.2021.00039

Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems. / Mitra, Alakananda; Mohanty, Saraju P.; Corcoran, Peter et al.
Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 149-154 (Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021).

Research output: Chapter in Book or Conference Publication/Proceeding › Conference Publication › peer-review

TY - GEN

T1 - Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems

AU - Mitra, Alakananda

AU - Mohanty, Saraju P.

AU - Corcoran, Peter

AU - Kougianos, Elias

PY - 2021

Y1 - 2021

N2 - Face Morphing has emerged as a pervasive attack of Facial Recognition Systems. The rapid growth of Generative Adversarial Networks takes it to a complete new level. Deepfake or deep neural network based face morphing, a.k.a deep-morph attack, presents a significant threat to Facial Recognition System. In this paper, we propose a novel Convolutional Neural Network based detection method of deep morphed deepfake images which is suitable for IoT environments in smart cities. A high accuracy of 94.83% has been achieved for the DeepfakeTIMIT HQ dataset. This lightweight and fast network is a natural choice for IoT environments.

AB - Face Morphing has emerged as a pervasive attack of Facial Recognition Systems. The rapid growth of Generative Adversarial Networks takes it to a complete new level. Deepfake or deep neural network based face morphing, a.k.a deep-morph attack, presents a significant threat to Facial Recognition System. In this paper, we propose a novel Convolutional Neural Network based detection method of deep morphed deepfake images which is suitable for IoT environments in smart cities. A high accuracy of 94.83% has been achieved for the DeepfakeTIMIT HQ dataset. This lightweight and fast network is a natural choice for IoT environments.

KW - Convolutional Neural Network

KW - Deep Learning

KW - Deep-fake

KW - Deep-Morph

KW - Facial recognition System

KW - Smart City

UR - https://www.scopus.com/pages/publications/85123339993

U2 - 10.1109/OCIT53463.2021.00039

DO - 10.1109/OCIT53463.2021.00039

M3 - Conference Publication

AN - SCOPUS:85123339993

T3 - Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021

SP - 149

EP - 154

BT - Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 19th Orissa Information Technology Society International Conference on Information Technology, OCIT 2021

Y2 - 16 December 2021 through 18 December 2021

ER -

Mitra A, Mohanty SP, Corcoran P, Kougianos E. Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems. In Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 149-154. (Proceedings - 2021 19th OITS International Conference on Information Technology, OCIT 2021). doi: 10.1109/OCIT53463.2021.00039

Detection of Deep-Morphed Deepfake Images to Make Robust Automatic Facial Recognition Systems

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this