Mapping of Facial Action Units to Virtual Avatar Blend Shape Movement

Simple item page
dc.contributor.authorWolff, Tonyen_US
dc.contributor.authorDollack, Felixen_US
dc.contributor.authorPerusquia-Hernandez, Monicaen_US
dc.contributor.authorUchiyama, Hideakien_US
dc.contributor.authorKiyokawa, Kiyoshien_US
dc.contributor.editorTanabe, Takeshien_US
dc.contributor.editorYem, Vibolen_US
dc.date.accessioned2024-11-29T06:38:12Z
dc.date.available2024-11-29T06:38:12Z
dc.date.issued2024
dc.description.abstractAction Units and blend shapes are two frameworks to describe facial movement. However, mappings between the two frameworks are underinvestigated. We present an automated mapping technique using machine learning. Our model infers ARKitcompatible blend shape weights from action unit intensities extracted with OpenFace. We use a GRU architecture to retain time-dependent information leveraging the particularities of Recurrent Neural Networks while still permitting fast, real-time inference. Our generalized model yields an activation precision of 90% and an activation recall of 85%.en_US
dc.description.sectionheadersPosters
dc.description.seriesinformationICAT-EGVE 2024 - International Conference on Artificial Reality and Telexistence and Eurographics Symposium on Virtual Environments - Posters and Demos
dc.identifier.doi10.2312/egve.20241392
dc.identifier.isbn978-3-03868-246-2
dc.identifier.issn1727-530X
dc.identifier.pages2 pages
dc.identifier.urihttps://doi.org/10.2312/egve.20241392
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/egve20241392
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Human-centered computing → Virtual reality; Computing methodologies → Machine learning algorithms
dc.subjectHuman centered computing → Virtual reality
dc.subjectComputing methodologies → Machine learning algorithms
dc.titleMapping of Facial Action Units to Virtual Avatar Blend Shape Movementen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
egve20241392.pdf
Size:
1.34 MB
Format:
Adobe Portable Document Format
Download
Collections
ICAT-EGVE2024 - Posters and Demos