We define a number of threat models to describe the goals, the available information and the actions characterising the behaviour of a possible attacker in multimedia forensic scenarios. We distinguish between an investigative scenario, wherein the forensic analysis is used to guide the investigative action and a use-in-court scenario, wherein forensic evidence must be defended during a lawsuit. We argue that the goals and actions of the attacker in these two cases are very different, thus exposing the forensic analyst to different challenges. Distinction is also made between model-based techniques and techniques based on machine learning, showing how in the latter case the necessity of defining a proper training set enriches the set of a...
Abstract Adversarial examples revealed the weakness of machine learning techniques in terms of robus...
Abstract—We introduce a theoretical framework in which to cast the source identification problem. Th...
Adversarial Machine Learning (AML) is emerging as a major eld aimed at the protection of automated ...
We define a number of threat models to describe the goals, the available information and the actions...
In recent decades, a significant research effort has been devoted to the development of forensic too...
The use of anti-forensic techniques is a very common practice that stealthy adversaries may deploy t...
Abstract—In the attempt to provide a mathematical back-ground to multimedia forensics, we introduce ...
The use of anti-forensic techniques is a very common practice that stealthy adversaries may deploy t...
In the attempt to provide a mathematical background to multimedia forensics, we introduce the source...
The existing solutions in the field of computer forensics are largely ad hoc. This paper discusses t...
This paper is a first attempt to provide a unified framework for studying signal processing problems...
This paper is a first attempt to provide a unified framework for studying signal processing problems...
Classification techniques are widely used in security settings in which data can be deliberately man...
The use of machine-learning for multimedia forensics is gaining more and more consensus, especially ...
Threat intervention with limited security resources is a chal-lenging problem. An optimal strategy i...
Abstract Adversarial examples revealed the weakness of machine learning techniques in terms of robus...
Abstract—We introduce a theoretical framework in which to cast the source identification problem. Th...
Adversarial Machine Learning (AML) is emerging as a major eld aimed at the protection of automated ...
We define a number of threat models to describe the goals, the available information and the actions...
In recent decades, a significant research effort has been devoted to the development of forensic too...
The use of anti-forensic techniques is a very common practice that stealthy adversaries may deploy t...
Abstract—In the attempt to provide a mathematical back-ground to multimedia forensics, we introduce ...
The use of anti-forensic techniques is a very common practice that stealthy adversaries may deploy t...
In the attempt to provide a mathematical background to multimedia forensics, we introduce the source...
The existing solutions in the field of computer forensics are largely ad hoc. This paper discusses t...
This paper is a first attempt to provide a unified framework for studying signal processing problems...
This paper is a first attempt to provide a unified framework for studying signal processing problems...
Classification techniques are widely used in security settings in which data can be deliberately man...
The use of machine-learning for multimedia forensics is gaining more and more consensus, especially ...
Threat intervention with limited security resources is a chal-lenging problem. An optimal strategy i...
Abstract Adversarial examples revealed the weakness of machine learning techniques in terms of robus...
Abstract—We introduce a theoretical framework in which to cast the source identification problem. Th...
Adversarial Machine Learning (AML) is emerging as a major eld aimed at the protection of automated ...