Detecting the camera model used to shoot a picture enables to solve a wide series of forensic problems, from copyright infringement to ownership attribution. For this reason, the forensic community has developed a set of camera model identification algorithms that exploit characteristic traces left on acquired images by the processing pipelines specific of each camera model. In this letter, we investigate a novel approach to solve camera model identification problem. Specifically, we propose a data-driven algorithm based on convolutional neural networks, which learns features characterizing each camera model directly from the acquired pictures. Results on a well-known dataset of 18 camera models show that: 1) the proposed method outperforms up-to-date state-of-the-art algorithms on classification of 64 × 64 color image patches; 2) features learned by the proposed network generalize to camera models never used for training.

The broad availability of tools for the acquisition and processing of multimedia signals has recently led to the concern that images and videos cannot be considered a trustworthy evidence, since they can be altered rather easily. This possibility raises the need to verify whether a multimedia content, which can be downloaded from the internet, acquired by a video surveillance system, or received by a digital TV broadcaster, is original or not. To cope with these issues, signal processing experts have been investigating effective video forensic strategies aimed at reconstructing the processing history of the video data under investigation and validating their origins. The key assumption of these techniques is that most alterations are not reversible and leave in the reconstructed signal some “footprints”, which can be analyzed in order to identify the previous processing steps. This paper presents an overview of the video forensic techniques that have been proposed in the literature, focusing on the acquisition, compression, and editing operations, trying to highlight strengths and weaknesses of each solution. It also provides a review of simple processing chains that combine different operations. Anti-forensic techniques are also considered to outline the current limitations and highlight the open research issues.

Due to the rapid proliferation of image capturing devices and user-friendly editing software suites, image manipulation is at everyone's hand. For this reason, the forensic community has developed a series of techniques to determine image authenticity. In this paper, we propose an algorithm for image tampering detection and localization, leveraging characteristic footprints left on images by different camera models. The rationale behind our algorithm is that all pixels of pristine images should be detected as being shot with a single device. Conversely, if a picture is obtained through image composition, traces of multiple devices can be detected. The proposed algorithm exploits a convolutional neural network (CNN) to extract characteristic camera model features from image patches. These features are then analyzed by means of iterative clustering techniques in order to detect whether an image has been forged, and localize the alien region.

The rise of AI-generated synthetic media, or deepfakes, has introduced unprecedented opportunities and challenges across various fields, including entertainment, cybersecurity, and digital communication. Using advanced frameworks such as Generative Adversarial Networks (GANs) and Diffusion Models (DMs), deepfakes are capable of producing highly realistic yet fabricated content, while these advancements enable creative and innovative applications, they also pose severe ethical, social, and security risks due to their potential misuse. The proliferation of deepfakes has triggered phenomena like “Impostor Bias”, a growing skepticism toward the authenticity of multimedia content, further complicating trust in digital interactions. This paper is mainly based on the description of a research project called FF4ALL (FF4ALL-Detection of Deep Fake Media and Life-Long Media Authentication) for the detection and authentication of deepfakes, focusing on areas such as forensic attribution, passive and active authentication, and detection in real-world scenarios. By exploring both the strengths and limitations of current methodologies, we highlight critical research gaps and propose directions for future advancements to ensure media integrity and trustworthiness in an era increasingly dominated by synthetic media.

The widespread use of deep learning techniques for creating realistic synthetic media, commonly known as deepfakes, poses a significant threat to individuals, organizations, and society. As the malicious use of these data could lead to unpleasant situations, it is becoming crucial to distinguish between authentic and fake media. Nonetheless, though deepfake generation systems can create convincing images and audio, they may struggle to maintain consistency across different data modalities, such as producing a realistic video sequence where both visual frames and speech are fake and consistent one with the other. Moreover, these systems may not accurately reproduce semantic and timely accurate aspects. All these elements can be exploited to perform a robust detection of fake content. In this paper, we propose a novel approach for detecting deepfake video sequences by leveraging data multimodality. Our method extracts audio-visual features from the input video over time and analyzes them using time-aware neural networks. We exploit both the video and audio modalities to leverage the inconsistencies between and within them, enhancing the final detection performance. The peculiarity of the proposed method is that we never train on multimodal deepfake data, but on disjoint monomodal datasets which contain visual-only or audio-only deepfakes. This frees us from leveraging multimodal datasets during training, which is desirable given their lack in the literature. Moreover, at test time, it allows to evaluate the robustness of our proposed detector on unseen multimodal deepfakes. We test different fusion techniques between data modalities and investigate which one leads to more robust predictions by the developed detectors. Our results indicate that a multimodal approach is more effective than a monomodal one, even if trained on disjoint monomodal datasets.