Experiment with machine learning with this super-smart standalone object and image recognition system!
Build your own intelligent camera that can see and recognize objects using machine learning. All of this fits in a handy little cardboard cube, powered by a Raspberry Pi.
The AIY Vision Kit includes all the components you need - including a Raspberry Pi Zero WH, a Raspberry Pi Camera v2 and all the necessary software (you'll just need to add a power supply). There are clear, step by step instructions that will show you how to assemble your AIY Vision Kit, connect to it, and run the Joy Detector demo, which recognizes faces and detects if they're smiling. You can then try running some other demos that detect other kinds of objects with the camera, and even install your own custom-trained TensorFlow model.
Google provides a fully documented Python API to code the inferencing ("recognition") of the images, and control the Vision Bonnet. The Vision Bonnet has some useful additional interfaces such as PWM, so it's great for adding image capture and processing capability to your Raspberry Pi based project.
Contents
- Vision bonnet
- Raspberry Pi Zero WH
- Raspberry Pi Camera v2
- Long flex cable
- Push button
- Button harness
- Micro USB cable
- Piezo buzzer
- Privacy LED
- Short flex cable
- Button nut
- Tripod nut
- LED bezel
- Standoffs
- MicroSD card
- Camera box cardboard
- Internal frame cardboard
Please note: the power supply is not included, but you can pick one up here.
Time required to build: 1.5 hours
Ages: 14+
TensorFlow models: MobileNetV1 / MobileNetV1 + SSD / SqueezeNet supported within certain constraints
Online Documentation
- Google Assembly instructions & main page about the AIY Vision kit
- Python API Library Overview
- Python API Library Documentation
- Vision Bonnet Documentation
- SD card software