Key result
High-quality deepfakes exhibit valid heart rates, challenging the assumption that they lack physiological signals.
Why the study?
Visual detection of deepfakes is increasingly difficult, motivating research into whether heart-beat-related signals from remote photoplethysmography can differentiate genuine videos from high-quality deepfakes.
High-quality deepfakes retain valid physiological heart rate signals from their driver videos, challenging the assumption that deepfakes lack valid physiological signals and complicating rPPG-based deepfake detection.
High-quality deepfakes may evade rPPG detection; leaves open need for robust validation before clinical video use.
Introduction Deepfakes have become ubiquitous in our modern society, with both their quantity and quality increasing. The current evolution of image generation techniques makes the detection of manipulated content through visual inspection increasingly difficult. This challenge has motivated researchers to analyze heart-beat-related signal to distinguish deep fakes from genuine videos. Methods In this study, we analyze deepfake videos of faces generated with novel methods regarding their heart-beat-related signals using remote photoplethysmography (rPPG). The rPPG signal describes the blood flow based, or rather local blood volume changes, and thus reflects the pulse signal. For our analysis, we present a pipeline that extracts rPPG signals and investigate the origin of the extracted signals in deepfake videos using correlation analyses. To validate our rPPG extraction pipeline and analyze rPPG signals of deepfakes, we captured a dataset of facial videos synchronized with an electrocardiogram (ECG) as a ground-truth pulse signal. Additionally, we generated high-quality deepfakes and incorporated publicly available datasets into our evaluation. Results We prove that our heart rate extraction pipeline produces valid estimates for genuine videos by comparing the estimated results with ECG reference data. Our high-quality deepfakes exhibit valid heart rates and their rPPG signals show a significant correlation with the corresponding driver video that was used to generate them. Furthermore, we show that this also holds for deepfakes from a publicly available dataset. Discussion Previous research assumed that the subtle heart-beat-related signals get lost during the deepfake generation process, making them useful for deepfake detection. However, this paper shows that this assumption is no longer valid for current deepfake methods. Nevertheless, preliminary experiments indicate that analyzing spatial distribution of bloodflow regarding its plausibility can still help to detect high quality deepfakes.
No takes yet. Share an insight, caveat, or question.
Seibold et al. (2025) studied Deepfake detection (n=12). rPPG-based heart rate extraction vs. Genuine videos was evaluated on Heart rate signal correlation between deepfakes and driver videos. High-quality deepfakes exhibit valid heart rates and their rPPG signals show a significant correlation with the corresponding driver video, challenging the assumption that deepfakes lack physiological signals.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: