This project is coming soon. Sign up to receive updates and be notified when this project launches.
DepthAI is a platform built around the Myriad X that gives neural inference coupled with depth vision and feature tracking. It’s a one-stop shop for folks who want to combine and harness the power of AI, depth, and tracking.
DepthAI - Raspberry Pi Carrier board prototype
Right now, the only way to solve this problem is to buy two different products (or in most cases, three) from different companies and then connect them together at the hardware level and software level yourself. This is a huge bummer if you care about power, size, or cost - or if you want it to just work out of the box.
The DepthAI motherboard is one solution that holds the Luxonis Myriad X Module and Raspberry Pi Compute Module.
A. OV9282 (onboard cameras, greyscale)
B. BW1099 Luxonis Myriad X module
C. IMX378 (onboard camera, color)
D. Connector for Raspberry Pi Compute Module 3B+
DepthAI was born out of the desire to improve bike safety. Using depth and AI (and feature tracking), we developed Commute Guardian, a artificial intellegence bike light that detects and prevents from-behind crashes. While working on this, we realized there isn’t a good embedded platform that allows for this. So we’re making it!
The Myriad X does depth, AI, and feature tracking, so it allows folks to embed it at low power and with a smaller solution that’s way easier to integrate into a design, manufacturing line, farm equipment, etc. And it leaves the included Raspberry Pi Compute 3B+ module at 0% CPU use. So not only is it cheaper, smaller, and easier to integrate. It leaves a home for your end-use code… the depth, AI, and feature tracking are all 100% on the Myriad X leaving the Raspberry Pi free for your code. Our solution is the first to do this.
Development model; not final version
Left; Intel Myriad X with calibration applied. Right; Myriad X with less depth smoothing.
Render: Myriad X with direct image sensors to Raspberry Pi 4
Render: Myriad X with direct image sensors to Raspberry Pi 3