.png){kind=icon}
.png){kind=icon}
Return
Laytex: A Data-Driven Tool for Automated and Customized Textile Sensor Layout Design in Motion Capture
.png){kind=icon}
PRE
AI
N
J
Y
L
Y
G
Y
Y
Q
T
DOI：10.1145/3749493.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Textile stretch sensors offer significant potential for motion tracking in human-computer interaction, yet designing precise sensor layouts remains challenging due to variable skin deformation patterns and manual design dependencies. We present Laytex, a data-driven tool that automates sensor layout generation through body deformation analysis and clustering-based optimization. The system processes 3D point cloud data to identify deformation hotspots, supports parametric customization of sensor quantities and lengths, and visualizes layouts with coverage reports. Evaluations with 10 participants performing shoulder motions demonstrated strong sensor-angle correlations (mean maximum coefficient: 0.76) and effective angles interpretation using LSTM networks (Mean Per-Joint Angular Error: 7.65°), comparable to state-of-the-art manually designed solutions. A workshop with 19 participants from diverse backgrounds further validated Laytex's cross-domain applicability and ability to streamline workflows, resulting in functional prototypes across applications. Laytex bridges the gap between computational design and practical deployment, offering a scalable solution for developing adaptive wearable technologies.
Journal
P
IF:
4.5
Papers:
1.1K
Citations:
7.2K
Organization
No organization information available
Cited Papers