Gravity: HUSKYLENS - An Easy-to-use AI Vision Sensor

HuskyLens is an easy-to-use AI vision sensor. It can learn to detect objects, faces, lines, colors and tags just by clicking.

Through the UART / I2C port, HuskyLens can connect to Arduino, Raspberry Pi, LattePanda, or micro:bit, and make your very creative projects without playing with complex algorithms.

HuskyLens aims to be the simplest AI vision sensor. You are able to change various algorithms by pressing buttons. By a simple click, HuskyLens will start learning new objects. After that, Huskylens can detect them. Now you can connect HuskyLens to the controller board, and make your projects.

HuskyLens is designed to be smart. Built-in algorithms allow HuskyLens to learn new things through a single click. Moreover, by long pressing, HuskyLens can continually learn them even from different perspectives and in various ranges. The more it learns, the smarter it is.

With the HuskyLens, your projects have new ways to interact with you or environment, such as interactive gesture control, autonomous robots, smart access control and interactive toy. There are so many new applications for you to explore.

HuskyLens can learn specific hand gestures, it recognizes learned hand movement patterns and feed its position. With these data, creating awesome interactive projects are never so easy.

HuskyLens can detect and follow lines. Line-following is not something new, there are plenty of methods and algorithms having excellent performance in this application. However, most of them require tedious parameters tuning. HuskyLens provides a new way to do line following, by simply clicking buttons, new lines can be learned and followed.

When combined with tag recognition, HuskyLens will automatically determine proper direction, and allow you to enjoy creating a self-driving robot.

HuskyLens can be eyes of robots. It allows your robot to recognize you, understand your hand gesture commands, help you organize your thing... Anything is possible!

SPECIFICATION

• Processor: Kendryte K210
• Image Sensor: OV2640 (2.0Megapixel Camera)
• Supply Voltage: 3.3~5.0V
• Current Consumption: 310mA@3.3V, 220mA@5.0V (face recognition mode; 80% backlight brightness; fill light off)
• Connection Interface: 

UART, 9600~115200bps

I2C, 100K~400K

• Display: 2.0-inch IPS screen with 320*240 resolution
• Built-in Algorithms: Object Tracking, Face Recognition, Object Recognition, Line Tracking, Color Recognition, Tag Recognition 
• Dimension: 52mm * 44.5mm