Gesture recognition and machine learning are getting a lot of air time these days, as people understand them more and begin to develop methods to implement them on many different platforms. Of course this allows easier access to people who can make use of the new tools beyond strictly academic or business environments. For example, rollerblading down the streets of Atlanta with a gesture-recognizing, streaming TV that [nate.damen] wears over his head.

He’s known as [atltvhead] and the TV he wears has a functional LED screen on the front. The whole setup reminds us a little of Deep Thought. The screen can display various animations which are controlled through Twitch chat as he streams his journeys around town. He wanted to add a little more interaction to the animations though and simplify his user interface, so he set up a gesture-sensing sleeve which can augment the animations based on how he’s moving his arm. He uses an Arduino in the arm sensor as well as a Raspberry Pi in the backpack to tie it all together, and he goes deep in the weeds explaining how to use Tensorflow to recognize the gestures. The video linked below shows a lot of his training runs for the machine learning system he used as well.

[nate.damen] didn’t stop at the cheerful TV head either. He also wears a backpack that displays uplifting messages to people as he passes them by on his rollerblades, not wanting to leave out those who don’t get to see him coming. We think this is a great uplifting project, and the amount of work that went into getting the gesture recognition machine learning algorithm right is impressive on its own. If you’re new to Tensorflow, though, we have featured some projects that can do reliable object recognition using little more than a Raspberry Pi and a camera.

The HackadayPrize2020 is Sponsored by:

Today when you get a text, you can respond with message via an on-screen keyboard. Looking into the future, however, how would you interact unobtrusively with a device that’s integrated into eyeglasses, contacts, or perhaps even something else?

TipText is one solution envisioned by researchers at Dartmouth College, which uses a MPR121 capacitive touch sensor wrapped around one’s index finger as a tiny 2×3 grid QWERTY keyboard.

The setup incorporates an Arduino to process inputs on the grid and propose a number of possible words on a wrist-mounted display that the user can select by swiping right with the thumb. A new word is automatically started when the next text entry tap is received, allowing for a typing speed of around 12-13 words per minute.

Good science fiction has sound scientific fact behind it and when Tony Stark first made his debut on the big screen with design tools that worked at the wave of a hand, makers and hackers were not far behind with DIY solutions. Over the years the ideas have become much more polished, as we can see with this Gesture Recognition with PIR sensors project.

The project uses the TPA81 8-pixel thermopile array which detects the change in heat levels from 8 adjacent points. An Arduino reads these temperature points over I2C and then a simple thresholding function is used to detect the movement of the fingers. These movements are then used to do a number of things including turn the volume up or down as shown in the image alongside.

The brilliant part is that the TPA81 8-Pixel sensor has been around for a number of years. It is a bit expensive though it has the ability to detect small thermal variations such as candle flames at up to 2 Meters. More recent parts such as the Panasonic AMG8834 that contain a grid of 8×8 such sensors are much more capable for your hacking/making pleasure, but come with an increased price tag.

This technique is not just limited to gestures, and can be used in Heat-Seeking Robots that can very well be trained to follow the cat around just to annoy it.