Deepfakes beyond the algorithm: a review of visual cues identifiable to the naked eye in the generative artificial intelligence era

• Camily Cantão UFPA
• Iago Medeiros UFPA

DOI: https://doi.org/10.5753/erin.2025.16056

Abstract

This article analyzes the visual detection of deepfake videos with the naked eye, focusing on human perception as a tool to aid in the increasingly sophisticated image and video generation techniques. It addresses the limitations of current automatic detectors and explores, based on the literature, recurring visual cues in faked videos, such as the absence of natural blinks, artificial expressions, exaggerated fluidity, homogeneous skin texture, and inconsistencies between the figure and background. A practical example with artificial intelligence (AI)-generated video complements the analysis. The findings reinforce that, despite high aesthetic realism, visual flaws can still be perceived by attentive observers. It concludes that critical and visual observation are viable avenues for combating audiovisual disinformation.

Keywords: Deepfake, Generative Artificial Intelligence, Observation

References

Diel, A., Lalgi, T., Schröter, I. C., MacDorman, K. F., Teufel, M., and Bäuerle, A. (2024). Human performance in detecting deepfakes: A systematic review and meta-analysis of 56 papers. Computers in Human Behavior Reports, 16:100538.

Guarnera, L. et al. (2020). A deepfake video detection method based on convolutional neural networks and frequency analysis. Sensors, 20(18):5112.

Gupta, P., Chugh, K., Dhall, A., and Subramanian, R. (2020). The eyes know it: Fakeet – an eye-tracking database to understand deepfake perception.

Josephs, E., Fosco, C., and Oliva, A. (2023). Artifact magnification on deepfake videos increases human detection and subjective confidence.

Korshunov, P. and Marcel, S. (2023). Survey of visual artifacts in deepfake videos. arXiv preprint arXiv:2311.10824.

Lee, A. (2019). How puny humans can spot devious deepfakes. [link]. Acesso em: Julho 2025.

Tolosana, R., Romero-Tapiador, S., Fierrez, J., and Vera-Rodriguez, R. (2020). Deepfakes evolution: Analysis of facial regions and fake detection performance.