Stacked Dual Attention for Joint Dependency Awareness in Pose Reconstruction and Motion Prediction

Simple item page
dc.contributor.authorGuinot, Lenaen_US
dc.contributor.authorMatsumoto, Ryutaroen_US
dc.contributor.authorIwata, Hiroyasuen_US
dc.contributor.editorJean-Marie Normanden_US
dc.contributor.editorMaki Sugimotoen_US
dc.contributor.editorVeronica Sundstedten_US
dc.date.accessioned2023-12-04T15:43:32Z
dc.date.available2023-12-04T15:43:32Z
dc.date.issued2023
dc.description.abstractHuman pose reconstruction and motion prediction in real-time environments have become pivotal areas of research, especially with the burgeoning applications in Virtual and Augmented Reality (VR/AR). This paper presents a novel deep neural network underpinned by a stacked dual attention mechanism, effectively leveraging data from just 6 Inertial Measurement Units (IMUs) to reconstruct human full body poses. While previous works have predominantly focused on image-based techniques, our approach, driven by the sparsity and versatility of sensors, taps into the potential of sensor-based motion data collection. Acknowledging the challenges posed by the under-constrained nature of IMU data and the inherent limitations in available open-source datasets, we innovatively transform motion capture data into an IMU-compatible format. Through a holistic understanding of joint dependencies and temporal dynamics, our method promises enhanced accuracy in motion prediction, even in uncontrolled environments typical of everyday scenarios. Benchmarking our model against prevailing methods, we underscore the superiority of our dual attention mechanism, setting a new benchmark for real-time motion prediction using minimalistic sensor arrangements.en_US
dc.description.sectionheadersLearning and 3D Reconstruction
dc.description.seriesinformationICAT-EGVE 2023 - International Conference on Artificial Reality and Telexistence and Eurographics Symposium on Virtual Environments
dc.identifier.doi10.2312/egve.20231326
dc.identifier.isbn978-3-03868-218-9
dc.identifier.issn1727-530X
dc.identifier.pages177-184
dc.identifier.pages8 pages
dc.identifier.urihttps://doi.org/10.2312/egve.20231326
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/egve20231326
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Computing methodologies → Real-time simulation; Motion processing; Reconstruction
dc.subjectComputing methodologies → Real
dc.subjecttime simulation
dc.subjectMotion processing
dc.subjectReconstruction
dc.titleStacked Dual Attention for Joint Dependency Awareness in Pose Reconstruction and Motion Predictionen_US
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
177-184.pdf
Size:
2.53 MB
Format:
Adobe Portable Document Format
Download
Collections
ICAT-EGVE2023