Looking at photos of dogs and cats can apparently make you feel happier, and self-proclaimed “depressed millennial” Ryder of the YouTube Channel Ryder Calm Down decided photos weren’t good enough and developed a camera that recognizes dogs and alerts him so he can spot them out the window.
As Gizmodo puts it, procrastinating by looking at real dogs is a lot better than procrastinating while looking at photos of dogs. Ryder’s system is built on the Raspberry Pi camera and a Raspberry Pi, which work together to analyze subjects using a pre-programmed machine learning system that is able to recognize about 80 different objects including people, cars, and dogs as demonstrated below.
Ryder injected a bit of custom code to get the camera operating to his specific desires which you can download here. When it’s fully set up and pointed out his front window, Ryder’s machine-learning-powered camera can recognize when a dog passes into frame and will alert him via an attached megaphone using an admittedly creepy robotic voice.
But what if Ryder wasn’t home? If they weren’t going to be there to appreciate the dogs, they thought that letting dog owners, and therefore the dogs, know that their animal looks nice would be another fun use of his creation.
“We can use the same technology to tell people that have a cool-looking dog even though we can’t really tell,” Ryder says. “So I turned the megaphone around and opened up the window.”
When the Raspberry Pi camera sees a dog, it can yell (in Ryder’s voice), “I like your dog!” The results are about as hilarious as you might expect, but a compliment is a compliment, right?
The screenshots above beautifully illustrate how unsettling it is to hear the voice shouting at you, as the person depicted looks around frantically in an attempt to figure out where the loud, disembodied voice is coming from, and why.
“I do feel better, but I’m not sure if it was the dogs or building something that actually worked for once,” Ryder concludes.
Whatever the case, what Ryder has built here is just the latest example that shows machine learning algorithms paired with cameras can have both practical uses as well as fun, if not somewhat silly, ones.