technology
health
scanning
WiPOSE
Human Pose Estimation for Smart Environments
Partners: SmartOpt (BE), TyX (TR), Almende (NL), IBSS (TR), TOBB University (TR), Vestel (TR)
Duration: 36 months (2025/28)
Budget: €3,300,000
Current Phase: Research & Prototyping
WiPOSE aims to develop a revolutionary system for human location and pose estimation using WiFi signals and deep learning techniques. It provides a privacy-preserving, cost-effective alternative to camera and wearable-based systems, addressing privacy concerns, high costs, and user inconvenience. The system is intended for applications in smart homes, elderly care, and industrial safety.
WiPOSE leverages WiFi signals and deep learning for human pose estimation, which eliminates the need for invasive cameras or wearable devices. The project uses WiFi signal-based sensing to detect poses in real-time, providing a non-invasive, cost-effective solution that ensures compliance with privacy regulations like GDPR. This is achieved by using advanced neural network models and electromagnetic simulations. The technology can be deployed in a range of applications including elderly care, workplace safety, and smart homes, where privacy and efficiency are critical. The integration of hardware accelerators like SoCs, FPGAs, and ASICs ensures that the system performs in real-time.
Stay tuned for more information, or inquire now.