My first crack at homebrew mo-cap. Sort of.
 

Not really mo-cap, but a fun project nonetheless. I'm working on the "real" glove now, which uses a metallic "exoskeleton" on a fabric glove, commodity potentiometers, and can also measure deflection between the fingers. So far, so good. I hope to post some pictures of that one in a couple of months.

YouTube - Cleaning Glove as Virtual Reality Input Device

Everything seen in the video is of my original design, including the circuit board, which was my first ever attempt to use all surface-mount electronics in a project. And boy! Was that ever a pain!!

PS: Don't ask me why I made the hand model green instead of blue. I'm sorry. :)

That is really neat. Must took along time to desgin and assemble! Great work, best of luck for your project. A++

Thanks! Going from idea to what's shown in the video took about five months. Happily, with the electronics mostly behind me, I've been able to confidently experiment with other (e.g., mechanical) aspects. The circuit board was designed to be the basis for a bunch of different input and biometrics projects, so it'll be a while before I have to crack the design on that. I hope!

This is awesome. Keep going!

Amazing :D

What's the design of the circuit board like?

Hi HNM:

The board is a USB bus-powered 14-channel sampler. Conceptually, it's nothing amazing. Its design is centered on a PIC 16F1934 microcontroller running at 32MHz. I chose that chip because of the large number of analog inputs it offered. The firmware basically just cycles through all these inputs very quickly--about 2500 samples/sec/channel with all channels in use--and hands the samples off to a FT245R USB FIFO which takes care of the USB. There is also a pair of 8-channel amplifiers between the inputs and the 16F1934 for conditioning, but they're not doing a lot for me in this glove application.

It works.

As I mentioned earlier, one of the bigger challenges for me was just the soldering. I've done plenty of electronics hacking, just not a lot with surface-mount stuff. Unfortunately, surface mount is what all the cool parts come in these days, so I needed to take the plunge at some point!

You've certainly done more than me there :P , I am impressed :)

Very nice. If I somehow end up a crazy billionaire, I shall be getting you to build me my nanosuit :P

I see your "Occupation: Student." You studying CS, CE, EE? I hope so. As somebody who does quite a bit of technical recruiting, I can say with certainty that the ranks of professional geekdom still need a LOT of help! :)

:D If only I could get paid to do this kind of stuff! Alas, a degree in electrical engineering is one you'll never see followed by "or equivalent experience" in a list of job qualifications. :P

CS, unfortunately. Turned out to not be particularly relevant or useful, although at least my main interests are still doable without a specific degree.

Ouch! Go easy on CS! :) I trust that "not ... relevant or useful" is meant to be taken in the context of your new goals. That's fair. You'll be glad to have the "piece of paper" though, so don't stop pursuing it! That's how employers deluged with resumes "cut the deck," so to speak.

Getting OT. Sorry.

But whatever you do decide to go after: Best of luck!

This is AWESOME! Great job! :D
A suggestion: you could add digital gyro, accelerometer and compass circuits, like the ones iPhone has for even more useful data :)

Definitely! I actually have several leftover inputs in this particular configuration to play with adding accelerometers and have been planning to do so. I want to finish off the deflection sensors for the fingers first, though. The rubber sensors I used here were an experiment and a partial failure. Their durability and accuracy suck, and I had to implement smoothing and other nasty hacks to cope with that, so they're out for the next round! :)

Why hadn't I seen this thread before :/

That's a cool glove :D

Pics please :D