Charmed Labs’ Machine Vision Camera Vizy Goes Live on Kickstarter

Charmed Labs unveiled its AI-driven Vizy camera last month, which takes advantage of onboard deep learning neural networks for object, people, and animal detection. Unlike other machine learning imaging systems that require a host PC, Vizy utilizes an internal Raspberry Pi 4 for processing and classification.

Beyond the Raspberry Pi, Vizy is outfitted with a Sony IMX219 image sensor, the same hardware found in the Pi Camera 2.1 module, which allows it to take advantage of the same drivers and software. The camera packs an accessible I/O header, an on/off power button, a USB Type-C port, two HDMI ports, and a stereo-out jack. Rounding out the specs are a pair of USB 2.0/3.0 ports on the back of the camera, an Ethernet port, a light mount, an RGB LED, a 12V power jack, and an integrated fan to keep the hardware cool.

Vizy features a web-based interface that's accessible from any mobile device or computer with a browser, and is programmed in Python with support for TensorFlow, OpenCV, PyTorch, and other libraries.

The AI camera is loaded with several ready-to-run projects, including a birdfeeder app that recognizes different species of birds as they feed on seeds and other foods, plus other animals looking to raid the fowl pantry. There's also Motionscope, a custom scientific app that tracks objects in motion and calculates their position, speed, and acceleration within each frame. Additionally, Vizy can be used to explore the night sky with motorized telescope mounts as well as to remotely monitor and feed pets.

Charmed Labs is currently crowdfunding Vizy on Kickstarter with pledges starting at $229, which gets you a 2Gb Raspberry Pi 4, 32Gb Ultra MicroSDXC card with all software pre-installed, a 12-megapixel camera, and a wide-angle lens. It even comes with a switchable IR filter for day or night use, I/O connectors and accessories, and a 25W AC power supply.