Wöhler L, Henningson JO, Castillo S, Magnor M (2020)
Publication Status: Published
Publication Type: Conference contribution
Publication year: 2020
Publisher: Springer Science and Business Media Deutschland GmbH
Series: Communications in Computer and Information Science
Book Volume: 1300
Pages Range: 120-127
Conference Proceedings Title: Computer Animation and Social Agents
ISBN: 9783030634254
DOI: 10.1007/978-3-030-63426-1_13
Videos obtained by current face swapping techniques can contain artifacts potentially detectable, yet unobtrusive to human observers. However, the perceptual differences between real and altered videos, as well as properties leading humans to classify a video as manipulated, are still unclear. Thus, to support the research on perceived realism and conveyed emotions in face swap videos, this paper introduces a high-resolution dataset providing the community with the necessary sophisticated stimuli.
Our recording process has been specifically designed to focus on human perception research and entails three scenarios (text-reading, emotion-triggering, and free-speech). We assess the perceived realness of our dataset through a series of experiments. The results indicate that our stimuli are overall convincing, even for long video sequences. Furthermore, we partially annotate the dataset with noticeable facial distortions and artifacts reported by participants.
APA:
Wöhler, L., Henningson, J.O., Castillo, S., & Magnor, M. (2020). PEFS: A Validated Dataset for Perceptual Experiments on Face Swap Portrait Videos. In Feng Tian, Xiaosong Yang, Daniel Thalmann, Weiwei Xu, Jian Jun Zhang, Nadia Magnenat Thalmann, Jian Chang (Eds.), Computer Animation and Social Agents (pp. 120-127). Bournemouth, GB: Springer Science and Business Media Deutschland GmbH.
MLA:
Wöhler, Leslie, et al. "PEFS: A Validated Dataset for Perceptual Experiments on Face Swap Portrait Videos." Proceedings of the 33rd International Conference on Computer Animation and Social Agents, CASA 2020, Bournemouth Ed. Feng Tian, Xiaosong Yang, Daniel Thalmann, Weiwei Xu, Jian Jun Zhang, Nadia Magnenat Thalmann, Jian Chang, Springer Science and Business Media Deutschland GmbH, 2020. 120-127.
BibTeX: Download