2024/05/14 by Arian Beckmann, Beckmann, Arian, Tilman Stephani +19
Computer Science · #Anomaly Detection Techniques and Applications #FOS: Computer and information sciences #FOS: Electrical engineering #Human-Computer Interaction (cs.HC) #Machine Learning (cs.LG) #Signal Processing (eess.SP) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2405.08527
openalex publication_date 2024/05/14 · openalex created_date 2025/11/01 · openalex updated_date 2026/07/28
Since the advent of Deepfakes in digital media, the development of robust and\nreliable detection mechanism is urgently called for. In this study, we explore\na novel approach to Deepfake detection by utilizing electroencephalography\n(EEG) measured from the neural processing of a human participant who viewed and\ncategorized Deepfake stimuli from the FaceForensics++ datset. These\nmeasurements serve as input features to a binary support vector classifier,\ntrained to discriminate between real and manipulated facial images. We examine\nwhether EEG data can inform Deepfake detection and also if it can provide a\ngeneralized representation capable of identifying Deepfakes beyond the training\ndomain. Our preliminary results indicate that human neural processing signals\ncan be successfully integrated into Deepfake detection frameworks and hint at\nthe potential for a generalized neural representation of artifacts in computer\ngenerated faces. Moreover, our study provides next steps towards the\nunderstanding of how digital realism is embedded in the human cognitive system,\npossibly enabling the development of more realistic digital avatars in the\nfuture.\n