Facial Features Matter: a Dynamic Watermark based Proactive Deepfake Detection Approach

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lan, Shulin, Liu, Kanlin, Zhao, Yazhou, Yang, Chen, Wang, Yingchao, Yao, Xingshan, Zhu, Liehuang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916492156600320
author Lan, Shulin
Liu, Kanlin
Zhao, Yazhou
Yang, Chen
Wang, Yingchao
Yao, Xingshan
Zhu, Liehuang
author_facet Lan, Shulin
Liu, Kanlin
Zhao, Yazhou
Yang, Chen
Wang, Yingchao
Yao, Xingshan
Zhu, Liehuang
contents Current passive deepfake face-swapping detection methods encounter significance bottlenecks in model generalization capabilities. Meanwhile, proactive detection methods often use fixed watermarks which lack a close relationship with the content they protect and are vulnerable to security risks. Dynamic watermarks based on facial features offer a promising solution, as these features provide unique identifiers. Therefore, this paper proposes a Facial Feature-based Proactive deepfake detection method (FaceProtect), which utilizes changes in facial characteristics during deepfake manipulation as a novel detection mechanism. We introduce a GAN-based One-way Dynamic Watermark Generating Mechanism (GODWGM) that uses 128-dimensional facial feature vectors as inputs. This method creates irreversible mappings from facial features to watermarks, enhancing protection against various reverse inference attacks. Additionally, we propose a Watermark-based Verification Strategy (WVS) that combines steganography with GODWGM, allowing simultaneous transmission of the benchmark watermark representing facial features within the image. Experimental results demonstrate that our proposed method maintains exceptional detection performance and exhibits high practicality on images altered by various deepfake techniques.
format Preprint
id arxiv_https___arxiv_org_abs_2411_14798
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Facial Features Matter: a Dynamic Watermark based Proactive Deepfake Detection Approach
Lan, Shulin
Liu, Kanlin
Zhao, Yazhou
Yang, Chen
Wang, Yingchao
Yao, Xingshan
Zhu, Liehuang
Computer Vision and Pattern Recognition
Cryptography and Security
Machine Learning
Image and Video Processing
Current passive deepfake face-swapping detection methods encounter significance bottlenecks in model generalization capabilities. Meanwhile, proactive detection methods often use fixed watermarks which lack a close relationship with the content they protect and are vulnerable to security risks. Dynamic watermarks based on facial features offer a promising solution, as these features provide unique identifiers. Therefore, this paper proposes a Facial Feature-based Proactive deepfake detection method (FaceProtect), which utilizes changes in facial characteristics during deepfake manipulation as a novel detection mechanism. We introduce a GAN-based One-way Dynamic Watermark Generating Mechanism (GODWGM) that uses 128-dimensional facial feature vectors as inputs. This method creates irreversible mappings from facial features to watermarks, enhancing protection against various reverse inference attacks. Additionally, we propose a Watermark-based Verification Strategy (WVS) that combines steganography with GODWGM, allowing simultaneous transmission of the benchmark watermark representing facial features within the image. Experimental results demonstrate that our proposed method maintains exceptional detection performance and exhibits high practicality on images altered by various deepfake techniques.
title Facial Features Matter: a Dynamic Watermark based Proactive Deepfake Detection Approach
topic Computer Vision and Pattern Recognition
Cryptography and Security
Machine Learning
Image and Video Processing
url https://arxiv.org/abs/2411.14798