Gravity: Offline Edge AI Gesture & Face Detection Sensor

The Gravity: Offline Edge AI Gesture & Face Detection Sensor transforms interaction into an intuitive experience using cutting-edge technology. With its integrated camera and onboard AI processor, this remarkable device identifies five predefined hand gestures and seamlessly detects up to 10 faces or upper bodies simultaneously within a 3-meter range. By fusing computer vision with real-time edge AI computing, it offers smooth, responsive interactions without the need for cloud connectivity, ensuring absolute privacy. Imagine effortlessly controlling smart devices or managing public interaction systems—all without touching a single button.

One of the most captivating features is its versatility in communication and power supply. The sensor is equipped with dual communication interfaces, I2C and UART, and supports a wide voltage range of 3.3V to 5V. This flexibility ensures compatibility with popular platforms like Arduino, Raspberry Pi, and ESP32, as well as graphical interfaces such as MakeCode and Mind+. Whether integrated into smart homes, touchless control systems, or privacy-focused IoT solutions, this module stands as a robust cornerstone for innovative ideas.

At the heart of its operation, the sensor's AI processor executes all recognition tasks locally, making sure no data is stored or transmitted online. This offline functionality assures data privacy—ideal for sensitive applications in healthcare, public spaces, or homes. The sensor operates on a mere 100mA, making it an excellent choice for projects where power efficiency is a priority. From contactless smart appliance control to interactive art installations and assistive devices, its low power consumption promises a perfect blend of performance and efficiency.

The sensor’s gesture recognition is precise, acknowledging five common gestures from a range of 0.5 to 3 meters. Each gesture—whether it’s a thumbs-up, OK, stop, victory, or call me gesture—triggers a corresponding colored RGB light, providing immediate visual feedback. For instance, a thumbs-up gesture lights up blue, while a victory sign activates a yellow hue, bringing interaction to life through vivid visuals.

Going beyond gestures, this advanced sensor excels in face and upper-body detection, identifying the head-and-shoulders area with ease. It can track up to 10 people simultaneously, outputting real-time data like position coordinates and confidence scores for the most clearly detected face. Such capabilities are valuable in applications like occupancy-based climate control, retail analytics for people-counting, or setting up a fan to follow a person’s movement across a space.

The sensor’s impressive 85° diagonal field of view paired with a 3-meter detection range ensures dependable performance across various settings. Whether it’s mounted on a wall, integrated into a device, or placed on a desk, its broad coverage adapts well to environments ranging from small rooms to larger, busy spaces like classrooms or lobbies. Such versatility makes the sensor a dynamic tool for evolving needs and innovative projects.