MoCapaci: Posture and gesture detection in loose garments using textile cables as capacitive
summary
A loose-fitting MoCaBlazer prototype embeds four textile antenna channels (Interactive Wear cables) with OpenTheremin V3 hardware to sense posture and gestures non-contact capacitive. Fourteen participants perform 20 upper-body gestures (400 instances each, one-size blazer, no per-user calibration); signals are bandpass 1–10 Hz and normalized. Deep models—a compact 1D neural model is preferred—reach ~97.18% leave-recording-out and 86.25% user-independent accuracy in an office with few metallic objects.
AI-written summary · about