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).

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to uniformly represent human characteristics and perform pose recognition through WiFi signals (specifically CSI, namely Channel State Information). The author proposes and constructs CSI - Net, which is a framework based on Deep Neural Network (DNN) for learning the representation of WiFi signals. Using CSI - Net, the author jointly solves two major human characteristic representation problems: biometric estimation (including body fat percentage, muscle percentage, water percentage and bone density percentage) and personal identification. In addition, the application of CSI - Net in two different pose recognition tasks is also demonstrated: hand motion recognition (fine - scale hand motions) and fall detection (coarse - scale body movements). The following is a more detailed description: 1. **Biometric Estimation**: Estimate physiological parameters such as body fat percentage, muscle percentage, water percentage and bone density percentage of the human body through WiFi signals. 2. **Personal Identification**: Use WiFi signals to identify the identities of different individuals. 3. **Hand Motion Recognition**: Recognize fine - scale hand motions, such as specific hand gestures. 4. **Fall Detection**: Detect coarse - scale body movements, especially fall events. The main contribution of the paper lies in proposing a unified deep - learning framework CSI - Net, which can handle multiple WiFi - sensing tasks simultaneously, and demonstrating its superior performance in different tasks in experiments. In addition, the author also constructs a database to support the research of these WiFi - sensing tasks.

CSI-Net: Unified Human Body Characterization and Pose Recognition

Skeleton-based human activity recognition with wifi CSI using a hybrid approach combining convolutional neural network and long short term memory

X-HRNet: Towards Lightweight Human Pose Estimation with Spatially Unidimensional Self-Attention

CSI-Former: Pay More Attention to Pose Estimation with WiFi

Modelling Human Body Pose for Action Recognition Using Deep Neural Networks

Simultaneous Implementation Features Extraction and Recognition Using C3D Network for WiFi-based Human Activity Recognition

DensePose From WiFi

Pose Estimation for Swimmers in Video Surveillance

Skeleton-Based Human Pose Recognition Using Channel State Information: A Survey

Classroom Student Posture Recognition Based on an Improved High-Resolution Network.

3D PostureNet: A unified framework for skeleton-based posture recognition

Human Activity Recognition based on WiFi Signal Using Deep Neural Network

CSI-F: A Human Motion Recognition Method Based on Channel-State-Information Signal Feature Fusion

Deep CSI Learning for Gait Biometric Sensing and Recognition

Seeing the Unseen: Wifi-based 2D Human Pose Estimation via an Evolving Attentive Spatial-Frequency Network

A novel approach for simultaneous human activity recognition and pose estimation via skeleton-based leveraging WiFi CSI with YOLOv8 and mediapipe frameworks

Robust Multi-User Pose Estimation Based on Spatial and Temporal Features from WiFi CSI

Towards 3D Human Pose Construction Using Wifi.

A Survey of Human Activity Recognition Using WiFi CSI

Motion Capture for Sporting Events Based on Graph Convolutional Neural Networks and Single Target Pose Estimation Algorithms

A Real-time Object Detection for WiFi CSI-based Multiple Human Activity Recognition