TruFor: Leveraging all-round clues for trustworthy image forgery detection and localization

被引:75
作者
Guillaro, Fabrizio [1 ]
Cozzolino, Davide [1 ]
Sud, Avneesh [2 ]
Dufour, Nicholas [2 ]
Verdoliva, Luisa [1 ]
机构
[1] Univ Federico II Naples, Naples, Italy
[2] Google Res, Mountain View, CA USA
来源
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2023年
关键词
STEGANALYSIS; CNN;
D O I
10.1109/CVPR52729.2023.01974
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this paper we present TruFor, a forensic framework that can be applied to a large variety of image manipulation methods, from classic cheapfakes to more recent manipulations based on deep learning. We rely on the extraction of both high-level and low-level traces through a transformer-based fusion architecture that combines the RGB image and a learned noise-sensitive fingerprint. The latter learns to embed the artifacts related to the camera internal and external processing by training only on real data in a self-supervised manner. Forgeries are detected as deviations from the expected regular pattern that characterizes each pristine image. Looking for anomalies makes the approach able to robustly detect a variety of local manipulations, ensuring generalization. In addition to a pixel-level localization map and a whole-image integrity score, our approach outputs a reliability map that highlights areas where localization predictions may be error-prone. This is particularly important in forensic applications in order to reduce false alarms and allow for a large scale analysis. Extensive experiments on several datasets show that our method is able to reliably detect and localize both cheapfakes and deepfakes manipulations outperforming state-of-the-art works. Code is publicly available at https://grip-unina.github.io/TruFor/.
引用
收藏
页码:20606 / 20615
页数:10
相关论文
共 47 条
[1]  
[Anonymous], 2016, ACM WORKSH INF HID M
[2]  
Avrahami Omri, 2022, ARXIV220602779
[3]  
Bammey Q, 2020, PROC CVPR IEEE, P14182, DOI 10.1109/CVPR42600.2020.01420
[4]  
Bi Xiuli, 2019, IEEE CVF C COMP VIS
[5]   Image Forgery Localization via Block-Grained Analysis of JPEG Artifacts [J].
Bianchi, Tiziano ;
Piva, Alessandro .
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2012, 7 (03) :1003-1017
[6]  
Bianchi T, 2011, INT CONF ACOUST SPEE, P2444
[7]  
Corbière C, 2019, ADV NEUR IN, V32
[8]   Noiseprint: A CNN-Based Camera Model Fingerprint [J].
Cozzolino, Davide ;
Verdoliva, Luisa .
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2020, 15 (01) :144-159
[9]  
Cozzolino D, 2015, IEEE INT WORKS INFOR
[10]  
Cozzolino Davide, 2022, DATA DRIVEN DIGITAL, P2