RuView, developed by ruvnet, is a groundbreaking system that leverages commodity WiFi signals to perform real-time human pose estimation, vital sign monitoring, and presence detection, all without the need for cameras, wearables, or an internet connection. This innovative approach, dubbed WiFi DensePose, analyzes Channel State Information (CSI) disturbances caused by human movement to reconstruct body position, breathing rate, and heartbeat. The system's core functionality is achieved through physics-based signal processing and machine learning algorithms, offering a privacy-focused solution for various applications.

The primary function of RuView is to "see through walls" using WiFi. It achieves this by utilizing CSI data, which provides detailed information about the amplitude and phase of WiFi signals as they travel through a space. By analyzing how these signals are altered by human presence and movement, the system can infer a wealth of information. Key features include real-time pose estimation, capable of tracking multiple individuals simultaneously; vital sign detection, including breathing and heart rate monitoring; and presence sensing with minimal latency. The system is designed to be multi-person, capable of tracking multiple individuals concurrently, and offers through-wall capabilities, allowing it to function even where cameras cannot.

The system's architecture is designed for efficiency and ease of use. Edge modules, small programs running directly on ESP32 sensors, eliminate the need for an internet connection and associated cloud fees, ensuring instant response times. The system is built using Rust, known for its performance and safety, and is highly optimized, achieving speeds of up to 54,000 frames per second for pose estimation. A key advantage is its one-command setup using Docker, enabling quick deployment and testing. The system also supports multi-arch Docker images, ensuring compatibility across different hardware platforms.

RuView's purpose is to provide a versatile and privacy-respecting sensing solution for a wide range of applications. It aims to replace or augment traditional sensing methods like cameras and wearables, particularly in scenarios where privacy is paramount or where environmental conditions make these methods impractical. The system's self-learning capabilities, powered by AI signal processing techniques, allow it to adapt to different environments without requiring extensive manual tuning or labeled data. The system's ability to work through walls, combined with its ability to detect vital signs, makes it particularly valuable in areas like healthcare, elder care, and disaster response.

The repository provides extensive documentation, including a user guide, build guide, and architecture decision records (ADRs), which detail the rationale behind design choices. The system's modular design, with edge modules for specific use cases, allows for customization and expansion. The system's applications span a broad spectrum, from smart home automation and retail analytics to robotics and industrial automation. It is particularly well-suited for situations where traditional sensing methods are limited by privacy concerns, environmental constraints, or cost. The system's ability to work with existing WiFi infrastructure and its low-cost hardware requirements make it a practical and scalable solution for various sensing needs.