New WHIP model reconstructs full-body motion from diverse wearable sensors

Researchers have developed a new method for reconstructing full-body motion from various wearable sensors, moving beyond fixed configurations. Their approach, named WHIP, utilizes a large-scale dataset that synchronizes consumer-grade devices like smartphones, smartwatches, and smart insoles with ground-truth 3D motion. This generative model can reconstruct motion from arbitrary subsets of available sensors, effectively handling missing modalities and producing physically plausible movements. AI

IMPACT Enables more flexible and realistic motion capture for applications using diverse wearable sensor setups.

RANK_REASON The cluster describes a new research paper published on arXiv detailing a novel generative model for motion reconstruction. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Andrea Boscolo Camiletto
• arXiv
• HMD-centric rigs
• IMU suits
• smart glasses
• smart insoles
• smartphones
• smartwatches
• WHIP

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →