DeepFake detection is a novel task for media forensics and is currently receiving a lot of research attention due to the threat these targeted video manipulations propose to the trust placed in video footage. The current trend in DeepFake detection is the application of neural networks to learn feature spaces that allow them to be distinguished from unmanipulated videos. In this paper, we discuss, with features hand-crafted by domain experts, an alternative to this trend. The main advantage that hand-crafted features have over learned features is their interpretability and the consequences this might have for plausibility validation for decisions made. Here, we discuss three sets of hand-crafted features and three different fusion strategies to implement DeepFake detection. Our tests on three pre-existing reference databases show detection performances that are under comparable test conditions (peak AUC > 0.95) to those of state-of-the-art methods using learned features. Furthermore, our approach shows a similar, if not better, generalization behavior than neural network-based methods in tests performed with different training and test sets. In addition to these pattern recognition considerations, first steps of a projection onto a data-centric examination approach for forensics process modeling are taken to increase the maturity of the present investigation.

In recent discussions in the European Parliament, the need for regulations for so-called high-risk artificial intelligence (AI) systems was identified, which are currently codified in the upcoming EU Artificial Intelligence Act (AIA) and approved by the European Parliament. The AIA is the first document to be turned into European Law. This initiative focuses on turning AI systems in decision support systems (human-in-the-loop and human-in-command), where the human operator remains in control of the system. While this supposedly solves accountability issues, it includes, on one hand, the necessary human–computer interaction as a potential new source of errors; on the other hand, it is potentially a very effective approach for decision interpretation and verification. This paper discusses the necessary requirements for high-risk AI systems once the AIA comes into force. Particular attention is paid to the opportunities and limitations that result from the decision support system and increasing the explainability of the system. This is illustrated using the example of the media forensic task of DeepFake detection.

Academic research in media forensics mainly focuses on methods for the detection of the traces or artefacts left by media manipulations in media objects. While the resulting detectors often achieve quite impressive detection performances, when tested under lab conditions, hardly any of those have yet come close to the ultimate benchmark for any forensic method, which would be courtroom readiness. This paper tries first to facilitate the different stakeholder perspectives in this field and then to partly address the apparent gap between the academic research community and the requirements imposed onto forensic practitioners. The intention is to facilitate the mutual understanding of these two classes of stakeholders and assist with first steps intended at closing this gap. To do so, first a concept for modelling media forensic investigation pipelines is derived from established guidelines. Then, the applicability of such modelling is illustrated on the example of a fusion-based media forensic investigation pipeline aimed at the detection of DeepFake videos using five exemplary detectors (hand-crafted, in one case neural network supported) and testing two different fusion operators. At the end of the paper, the benefits of such a planned realisation of AI-based investigation methods are discussed and generalising effects are mapped out.