Schlagwort: facial recognition

  • Raspberry Pi won’t let your watched pot boil

    Raspberry Pi won’t let your watched pot boil

    Reading Time: 3 minutes

    One of our favourite YouTubers, Harrison McIntyre, decided to make the aphorism “a watched pot never boils” into reality. They modified a tabletop burner with a Raspberry Pi so that it will turn itself off if anyone looks at it.

    [youtube https://www.youtube.com/watch?v=F1XSx-eaCAY?feature=oembed&w=500&h=281]

    In this project, the Raspberry Pi runs facial detection using a USB camera. If the Raspberry Pi finds a face, it deactivates the burner, and vice versa.

    There’s a snag, in that the burner runs off 120 V AC and the Raspberry Pi runs off 5 V DC, so you can’t just power the burner through the Raspberry Pi. Harrison got round this problem using a relay switch, and beautifully explains how a relay manages to turn a circuit off and on without directly interfacing with the circuit at the two minute mark of this video.

    The Raspberry Pi working through the switchable plug with the burner

    Harrison sourced a switchable plug bar which uses a relay to turn its own switches on and off. Plug the burner and the Raspberry Pi into that and, hey presto, you’ve got them working together via a relay.

    The six camera setup

    Things get jazzy at the four minute 30 second mark. At this point, Harrison decides to upgrade his single camera situation, and rig up six USB cameras to make sure that no matter where you are when you you look at the burner, the Raspberry Pi will always see your face and switch it off.

    Inside the switchable plug

    Harrison’s multiple-camera setup proved a little much for the Raspberry Pi 3B he had to hand for this project, so he goes on to explain how he got a bit of extra processing power using a different desktop and an Arduino. He recommends going for a Raspberry Pi 4 if you want to try this at home.

    Kit list:

    • Raspberry Pi 4
    • Tabletop burner
    • USB cameras or rotating camera
    • Switchable plug bar
    • All of this software
    It’s not just a saying anymore, thanks to Harrison

    And the last great thing about this project is that you could invert the process to create a safety mechanism, meaning you wouldn’t be able to wander away from your cooking and leave things to burn.

    We also endorse Harrison’s advice to try this with an electric burner and most definitely not a gas one; those things like to go boom if you don’t play with them properly.

    Website: LINK

  • Your face, 14 ft tall: image mapping with As We Are

    Your face, 14 ft tall: image mapping with As We Are

    Reading Time: 3 minutes

    While at World Maker Faire New York last weekend, I found myself chatting to a rather lovely gentleman by the name of Mac Pierce. During our conversation, Mac mentioned a project he’d worked on called As We Are, an interactive art installation located in the Greater Columbus Convention Center in Columbus, Ohio.

    as we are

    “So it’s this 14-foot head covered in LEDs…”, Mac began, and after his brief explanation, I found myself grabbing nearby makers to have him tell them about the project too. I was hooked! I hadn’t even seen photos of the sculpture, yet I was hooked. And true to his word, Mac had the press release for As We Are sitting in my inbox when I returned to Pi Towers.

    So here is it:

    The Greater Columbus Convention Center: “As We Are” – Creating the Ultimate Selfie Machine

    DCL, an award-winning fabricator of architectural specialties and custom experiential design elements, worked with artist Matthew Mohr to develop, engineer and fabricate this 14ft, 7,000lb, interactive digital sculpture. Featuring custom LED modules, an integrated 3D photobooth, 32 cameras, and a touch-screen display – this unique project combines technologies to present a seamless experience for visitors to display their own portrait on the sculpture.

    As We Are

    The brainchild of artist Matthew Mohr, As We Are was engineered and produced by DCL, an award-winning Boston-based fabricator whose greatest achievement to date, in my opinion at least, is hiring Mac Pierce.

    as we are

    YAY!

    DCL built the 14-foot structure using 24 layers of aluminium ‘ribs’ covered in custom Sansi LED modules. These modules add up to an astounding 850000 individual LEDs, allowing for crisp detail of images displayed by the build.

    as we are

    When a visitor to the Convention Center steps inside the interactive sculpture, they’re met with a wall of 32 Raspberry Pis plus Camera Modules. The Pis use facial recognition software to 3D scan the visitor’s face and flattened the image, and then map the face across the outer surface of the structure.

    Matthew Mohr was inspired to show off the diversity of Columbus, OH, while also creating a sense of oneness with As We Are. Combining technology and interaction, the sculpture has been called “the ultimate selfie machine”.

    If you’re in or near Columbus and able to visit the installation, we’d love to see your photos, so please share them with us on our social media platforms.

    Raspberry Pi facial mapping as we are

    You see now why I was dumbstruck when Mac told me about this project, yes?

    Always tell us

    Had it not been for a chance encounter with Mac at Maker Faire, we may never have heard of As We Are. While Matthew Mohr and DCL installed the sculpture in 2017, very little fuss was made about the use of Raspberry Pis within it, and it completely slipped under our radar. So if you are working on a project for your business, as a maker, or for any other reason, and you’re using a Raspberry Pi, please make sure to let us know by emailing [email protected].

    Website: LINK

  • facepunch: the facial recognition punch clock

    facepunch: the facial recognition punch clock

    Reading Time: 2 minutes

    Get on board with facial recognition and clock your screen time with facepunch, the facial recognition punch clock from dekuNukem.

    dekuNukem facepunch raspberry pi facial recognition

    image c/o dekuNukem

    How it works

    dekuNukem uses a Raspberry Pi 3, the Raspberry Pi camera module, and an OLED screen for the build. You don’t strictly need to include the OLED board, but it definitely adds to the overall effect, letting you view your daily and weekly screen time at a glance without having to access your Raspberry Pi for data.

    As dekuNukem explains in the GitHub repo for the build, they used a perf board to mount the screen and attached it to the Raspberry Pi. This is a nice, simple means of pulling the whole project together without loose wires or the need for a modified case.

    dekuNukem facepunch raspberry pi facial recognition

    image c/o dekuNukem

    This face_recognition library lets the Pi + camera register your face. You’ll also need a well lit 400×400 photograph of yourself to act as a reference for the library. From there, a few commands should get you started.

    Uses for facial recognition

    You could simply use facepunch for its intended purpose, but here at Pi Towers we’ve been discussing further uses for the build. We’re all guilty of sitting for too long at our desks, so why not incorporate a “get up and walk around” notification? How about a flashing LED that tells you to “drink some water”? You could even go a little deeper (though possibly a little Big Brother) and set up an “I’m back at my desk” notification on Slack, to let your colleagues know you’re available.

    You could also take this foray into facial recognition and incorporate it into home automation projects: a user-identifying Magic Mirror, perhaps, or a doorbell that recognises friends and family.

    What would you do with facial recognition on a Raspberry Pi?

    Website: LINK