A joint Disney Research and CMU team have produced a demo showing gesture controls on a variety of everyday, non-computer objects. The system, called Touché, uses capacitive coupling to infer things about what your hands are doing. It can determine which utensil you're eating your food with, or how you're grasping a doorknob, or even whether you're touching one finger to another or clasping your hands together. It's a pretty exciting demo, and the user interface possibilities are certainly provocative. Here's some commentary from Wired UK's Mark Brown:
Some of the proof-of-concept applications in the lab include a smart doorknob that knows whether it has been grasped, touched, or pinched; a chair that dims the lights when you recline into it; a table that knows if you're resting one hand, two hands, or your elbows on it; and a tablet that can be pinched from back to front to open an on-screen menu.
The technology can also be shoved in wristbands, so you can make sign-language-style gestures to control the phone in your pocket—two fingers on your palm to change a song, say, or a clap to stop the music. It can also go in liquids, to detect when fingers and hands are submerged in water.
"In our laboratory experiments, Touché demonstrated recognition rates approaching 100 percent," claims Ivan Poupyrev, senior research scientist at Disney Research in Pittsburgh. "That suggests it could immediately be used to create new and exciting ways for people to interact with objects and the world at large."
Disney researchers put gesture recognition in door knobs, chairs, fish tanks
The Nightmare Machine is an MIT project to use machine learning image-processing to make imagery for Hallowe’en.
The Stormtrooper Decanter is on back-order, but you can pre-order one from the next batch for £22 — it’s based on Andrew Ainsworth’s original movie helmet moulds from 1976, and will provide endless opportunities to point to lowball glasses and say things like “aren’t you a little short for a Stormtrooper drink?” (via Bonnie Burton)
Yahoo has released a machine-learning model called open_nsfw that is designed to distinguish not-safe-for-work images from worksafe ones. By tweaking the model and combining it with places-CNN, MIT’s scene-recognition model, Gabriel Goh created a bunch of machine-generated scenes that score high for both models — things that aren’t porn, but look porny.
Geek Fuel is a subscription delivery service that caters to those of us that love comics, gaming, and general geek culture. Every month, Geek Fuel will assemble a box of goodies with a value of $50 or over. The specific items are a mystery, but you’ll always get an exclusive t-shirt not found anywhere else, a full […]
If you like to DIY and you like helicopters, you’re going to really love the Flexbot Hexacopter Kit. This copter blows traditional models out of the water: it includes everything you need to actually build your own hexacopter, and then pilot it like a pro, too.The construction is complicated enough to give you a challenge, […]
This week’s top deals from the Boing Boing Store range from lobster to wine to desk organization. 1. Get Maine Lobster (50% Off)With these discounted packages from Get Maine Lobster, you can experience the sweet, fresh flavor of world-renowned Maine lobster right at your own dinner table. There are four options to choose from, each at […]