CSI-Net: Unified Human Body Characterization and Pose Recognition

TL;DR

CSI-Net is a deep learning framework that leverages WiFi signals for comprehensive human body analysis, including biometrics, recognition, and pose detection, demonstrating versatile applications in health and security domains.

Contribution

The paper introduces CSI-Net, a unified neural network that simultaneously performs body characterization and pose recognition using WiFi signals, a novel integration of these tasks.

Findings

CSI-Net accurately estimates body composition metrics.

It effectively recognizes individuals and hand signs.

It detects falls with high reliability.

Abstract

We build CSI-Net, a unified Deep Neural Network~(DNN), to learn the representation of WiFi signals. Using CSI-Net, we jointly solved two body characterization problems: biometrics estimation (including body fat, muscle, water, and bone rates) and person recognition. We also demonstrated the application of CSI-Net on two distinctive pose recognition tasks: the hand sign recognition (fine-scaled action of the hand) and falling detection (coarse-scaled motion of the body).