The things people can do with Raspberry Pis, especially when it comes to photography and video related projects, always fascinates me. And Adafruit has recently posted a very interesting project. It’s a camera that actually knows what it’s looking at. It’s based around Adafruit’s own BrainCraft HAT system, which is an AI Machine Learning addon for the Raspberry Pi 4. This project shows it in practical use in an actual project.
The camera uses TensorFlow Lite object recognition software to be able to understand what it’s seeing. And it’s compatible with both the 8-megapixel Pi Camera Module and the newer 12.3-megapixel interchangeable lens module. A 3D printed enclosure wraps everything up into a nice compact package to be able to take it out with you on the go.
While just being able to see and recognise an object in front of the camera might not be all that useful on its own, it opens up a lot of options for connecting the Pi to a “real camera” for shooting photos or video. You could connect a DSLR or mirrorless camera from its trigger port into the Pi’s GPIO pins, or even use a USB connection with something like gPhoto, to have it shoot a photo or start recording video when it detects a specific thing enter the frame.
Maybe you could use it as a remote wildlife monitoring camera that only fires when it sees a specific species of animal. Or maybe you can have it fire the camera and automatically inject keyword metadata into the image files as it transfers them off the camera and onto the Pi’s own internal storage.
As you can see from the video, though, it doesn’t always quite get things perfect. But, it’s a learning computer. So, it’ll figure it out eventually.
Whatever you ultimately choose to do with it, it’s a very cool little project. I’d be curious to see if many photographers have a go at making one of these and how they incorporate it into some real shoots.
[via Adafruit]
FIND THIS INTERESTING? SHARE IT WITH YOUR FRIENDS!