🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Interfacing with our smart insoles on an Apple TV using our Swift package + SwiftUI sample app

We’ve merged with BrilliantSole - follow them for future Ukaton updates

Experimenting with simple wrist controls using our Ukaton hardware, all running in a webpage A-Frame Their hardware is modular, so we can remove the insole & shoe clip, add a loop clip, and strap it around our wrist, relying just on motion+haptics We used

Virtual punching bag, using our Ukaton hardware to detect punches and provide haptic feedback Running an ML model on-device to detect punches, hooks, and uppercuts, made with Edge Impulse All running in the browser A-Frame

Adding haptic feedback to create immersive WebXR experiences on the Quest 3 A-Frame Here we vibrate our Ukaton modules differently based on hand gesture (pet, punch, grab, and release) We added WiFi support to our JavaScript SDK, allowing us to connect to our

Stomp, Kick, and Toss green shells in WebXR on the Quest 3 A-Frame By strapping our Ukaton modules to our ankle, we can detect kicks and stomps by running an Edge Impulse model on-device, and then send the results to the headset via WebSockets We also add haptic

Kicking a virtual soccer ball in WebXR on the Quest 3 ⚽️ A-Frame Using our Ukaton motion module (strapped to our ankle) to detect kicks and stomps

Ported our firmware to the Brilliant Labs Frame 👓, making it compatible with our SDK’s

Streaming camera data from smart glasses 👓 to a smart watch ⌚️ via Bluetooth As we port our smart insole firmware to the Brilliant Labs Frame, we added camera support to our SDK’s For context, this is running on an Apple Watch Series 5 (2019)

I posted a small description on Quest browser's new experimental "shared spaces" feature: github.com/cabanier/share… It automatically creates a shared coordinate space for headsets in the same room. Here's a small recording that shows it in action:

Tongue-controlled phone Here’s a selfie I took with my tongue

Track hands, faces, and objects with your smart glasses 👓 Running MediaPipe and Hugging Face models in the browser, using camera data sent from the Brilliant Labs Frame via Bluetooth The Frame is running custom firmware ported from our smart insoles, using our own

Running speech recognition on smart glasses 👓 As we port our smart insole firmware to the Brilliant Labs Frame, we added microphone support to our SDK’s, allowing us to easily run Hugging Face’s Whisper Web to perform speech recognition in the browser (both realtime and

Added display support for smart glasses (in this case the Brilliant Labs Frame) to our SDK Easily draw primitives (rectangles, rounded rectangles, circles, ellipses, polygons, line segments), set outline thickness, change colors, apply rotations, and even apply regular and

Added a smart glasses display simulator to our JavaScript SDK Easily preview how your applications will look in the browser, using your webcam, a video, or an A-Frame WebXR scene as the background Even works alongside compatible smart glasses in realtime (in this case the

Our smart glasses display simulator also works in WebXR, allowing us to preview AR Glasses applications on headsets like the Quest 3 in A-Frame And despite Web Bluetooth not working in the oculus browser, our JavaScript SDK allows us to indirectly connect to compatible smart

Express yourself with smart glasses 👓 Using MediaPipe’s face tracking to visualize a user’s eyes and eyebrows on a smart glasses display (in this case the Brilliant Labs Frame running custom firmware), sorta like a simple FaceTime Memoji While Frame doesn’t have built-in

Smart Glasses Webcam Used OBS to create a Virtual Webcam of our smart glasses camera stream demo Use in video conference apps like Zoom and Google Meet, for live-streaming on Twitch, or even as a quick way to test realtime computer vision demos (eg Hugging Face spaces)