It is currently Mon Sep 23, 2019 1:28 am



Reply to topic  [ 37 posts ] 
 DIY data gloves? 
Author Message
One Eyed Hopeful

Joined: Wed Sep 12, 2012 10:22 am
Posts: 8
Reply with quote
Watched the movie on this thread and started thinking...
http://www.mtbs3d.com/phpBB/viewtopic.php?f=140&t=15471&hilit=robot+game

Idea: Take a Razer Hydra, remove the sensors, attach them to 2 tight gloves with a hard backhand surface (so they don't flex so much when you move).
Attach something like these to the overside of the fingers: https://www.sparkfun.com/products/8606
Read each fingers bend value with a ┬ÁC and send the data over USB.

What do you think?


Mon Oct 01, 2012 9:02 pm
Profile
Petrif-Eyed
User avatar

Joined: Sat Sep 17, 2011 9:23 pm
Posts: 2220
Location: Menlo Park, CA
Reply with quote
I say go for it.


Mon Oct 01, 2012 9:23 pm
Profile
Certif-Eyable!
User avatar

Joined: Sat Dec 22, 2007 3:38 am
Posts: 990
Reply with quote
Sounds fantastic to me! I'm not ready to pull my hydra apart yet, but please post progress if you do. :D


Mon Oct 01, 2012 9:56 pm
Profile
Certif-Eyable!

Joined: Tue Sep 18, 2012 10:32 pm
Posts: 1139
Reply with quote
Removing the sensors might be harder than you think.

And doing anything useful with finger bend information (without even readings for each joint) is more difficult than you think without haptic feedback.


Mon Oct 01, 2012 9:58 pm
Profile
Certif-Eyable!
User avatar

Joined: Sat Dec 22, 2007 3:38 am
Posts: 990
Reply with quote
Regarding the flex sensor, I did a youtube search... maybe you should talk to the guy in this video, he looks familiar:
http://www.youtube.com/watch?v=jQXTV2hiNUQ


Mon Oct 01, 2012 9:59 pm
Profile
Golden Eyed Wiseman! (or woman!)

Joined: Fri Jul 08, 2011 11:47 pm
Posts: 1498
Reply with quote
Easily done, did something quite similar with my P5 Glove.

I'd recommend a P5 glove for this actually. Unfortunately, mine has stuffed out, so am about to buy another. It already gives you finger bend info, and combined with the Hydra is a pretty good solution... P5 Gloves are very cheap as well.

Those bend sensors look interesting, although in the comments it states you have to modify them to get any sort of decent life out of them (and some people had issues after modding them). However, those or a similar thing would be sweet for a DIY glove (I just dont see the point with the P5 glove being available already however).

@ 2EyeGuy: not sure why you say you need haptic feedback. With bend info from a P5, you can easily click, lock/unlock the cursor, and do other stuff. Most VR gloves dont have haptic feedback and work fine? Also, re removing the sensors, this can be tricky, but is doable. There are 3 basic methods, removing the coils and extending them, removing the joystick/buttons and trimming it down to just the 2 circult boards (which you can stack on each other, giving a 2-3" long sensor), or, possibly, removing the button board and using the unconfirmed mod posted in the Hydra tear down to turn it on.

The handy thing about the Hydra, is that it already has mouse control built in. My intention before my P5 died was to map the finger bend to the mouse button, so I can 'click' the mouse etc while playing Black Shark. Actually, thinking about it, I should probably investigate directly reading the P5 gloves bend data off the glove port, before I look at buying a new one...


Mon Oct 01, 2012 10:15 pm
Profile
Certif-Eyable!

Joined: Tue Sep 18, 2012 10:32 pm
Posts: 1139
Reply with quote
Since when do Virtual Reality gloves work at all for VR? Have you ever heard of anyone using one (pinch gloves don't count)? People use wands, and controllers, and gun models.

Don't get me wrong, I want to use VR gloves with my Rift. But naturally interacting with virtual objects using your hands isn't going to work as well as you imagine.


Mon Oct 01, 2012 10:38 pm
Profile
One Eyed Hopeful

Joined: Wed Sep 12, 2012 10:22 am
Posts: 8
Reply with quote
P5 sounds cool, perhaps mod one of these instead is a better way to go!

Of course you have to make a custom physics simulation for it to work as it should.


Mon Oct 01, 2012 10:46 pm
Profile
Golden Eyed Wiseman! (or woman!)

Joined: Fri Jul 08, 2011 11:47 pm
Posts: 1498
Reply with quote
VR gloves have worked for a long time, I believe there are still a few commercial ones on sale.
Back in the heyday of VR they were an essential. However, I agree that they are not absolutely necessary for modern VR (whatever that is at this point). However, I believe they add significantly to the immersion (being able to see your fingers wiggle etc helps with this). Also, there are many VR actions that cannot be performed as well with wands, Hydras, etc. For example, opening a drawer, throwing a grenade (dont let go of that Hydra!) etc - all of these are much more natural with a glove.

In the past, I believe most gestures were fairly basic. For example, to pick up an object, you made a fist. To drop it, you just open your hand.

Its not the same as reality, but its a step up from wands etc.

The P5 glove has a few basic physics apps, like one where you can pick up bricks and blocks etc with a 3D hand representation.


Mon Oct 01, 2012 11:30 pm
Profile
One Eyed Hopeful

Joined: Wed Sep 12, 2012 10:22 am
Posts: 8
Reply with quote
Ugly as hell but seem fun ;)
http://www.youtube.com/watch?v=3NVldkaDWy8


Mon Oct 01, 2012 11:49 pm
Profile
One Eyed Hopeful

Joined: Mon Sep 17, 2012 6:27 pm
Posts: 18
Reply with quote
In another thread, someone suggested mounting a LeapMotion sensor on the HMD for hand tracking. $70 pre-order, 8 cu. ft. of tracking volume with reportedly very good sensitivity and rate.


Tue Oct 02, 2012 12:18 am
Profile
Certif-Eyable!

Joined: Tue Sep 18, 2012 10:32 pm
Posts: 1139
Reply with quote
In real life, opening a drawer or throwing a grenade would be difficult without haptic feedback, and when your hand can go through solid objects.

It would have to be specially coded to allow your virtual hand and fingers to move separately from your real hand when there are physics constraints (such as holding the drawer handle) but spring back (and let go of the handle) when you move your real hand too far from where the virtual one can go.

And holding things would have to be specially coded rather than just using generic physics.

I've tried to implement picking up bricks and blocks and pushing or flicking them around before, and it's not as easy as you'd think. Because there's no way of measuring force.

Also, without any buttons or joysticks it is hard to move around and perform game actions.

I think you should still go for it and build it if you want to, but it might be slightly disappointing until people have worked out the best ways of coding for it.


Tue Oct 02, 2012 12:23 am
Profile
Golden Eyed Wiseman! (or woman!)

Joined: Fri Jul 08, 2011 11:47 pm
Posts: 1498
Reply with quote
Force is mass X acceleration. Therefore, virtual force depends entirely on the 'weight' of the virtual object. There is no problem with this. Using my grenade example, you can calculate the distance and flightpath, based on its weight, and the speed of the hand movement at point of release.

Stuff like spring back physics, constraints, etc, is 'nice' but not essential to use of a glove. I know what you mean in that it is clumsy to manipulate objects if you want to push them around realistically, but in terms of opening drawers and doors, picking up objects, etc, its better IMHO than using a wand and buttons, which breaks immersion more.

Movement with a glove is not that hard, you point with a finger in the direction you want to go, and the angle of your thumb controls speed. Gloves were the main input device for the old school VR's such as VPL's systems.


Tue Oct 02, 2012 4:30 pm
Profile
Two Eyed Hopeful
User avatar

Joined: Sat Jul 07, 2012 7:03 pm
Posts: 74
Reply with quote
This is the future (0:51):
http://www.youtube.com/watch?v=PidqxAAjUl4#t=51s


Tue Oct 02, 2012 5:59 pm
Profile WWW
Sharp Eyed Eagle!
User avatar

Joined: Tue Feb 21, 2012 11:57 pm
Posts: 429
Location: Irvine, CA
Reply with quote
On the subject of the Leap Motion sensor, I recently came across this thread on Leap Motion's forum.

Relevant bit from the co-founder:
Quote:
The Leap will have a field of view larger than the Oculus Rift HMD. So even a single one should work pretty well.


Tue Oct 02, 2012 7:18 pm
Profile
Certif-Eyable!
User avatar

Joined: Sat Dec 22, 2007 3:38 am
Posts: 990
Reply with quote
FingerFlinger wrote:
On the subject of the Leap Motion sensor, I recently came across this thread on Leap Motion's forum.

Relevant bit from the co-founder:
Quote:
The Leap will have a field of view larger than the Oculus Rift HMD. So even a single one should work pretty well.

That really could be the way to go.
I wonder if there is a way to have a formable object that could create and vary shapes for actual feedback. I'm imagining something like a series of balloons under your fingers that would inflate to a certain size depending on what was there. Somewhat complex, I guess.


Tue Oct 02, 2012 8:09 pm
Profile
Golden Eyed Wiseman! (or woman!)

Joined: Fri Jul 08, 2011 11:47 pm
Posts: 1498
Reply with quote
There have actually been quite a few force feedback gloves etc.
There was a british one that used air bladders a bit like you suggest, although I think these were mainly on the palm. There are also exoskeleton type ones. One issue with air, is that you then need an air hose going to your glove...


Tue Oct 02, 2012 8:22 pm
Profile
3D Angel Eyes (Moderator)
User avatar

Joined: Sat Apr 12, 2008 8:18 pm
Posts: 11394
Reply with quote
Well full body tracking, especially of the hands and fingers, is probably the holy-grail of VR HCI, so I don't know why anyone would say it can't work. Really its just a software issue at this point. There is nothing wrong with the concept. I plan to explore this more in the future with a really compelling demo but I've gotten busy with the 3D driver I'm working on. Currently I think using the Razer Hydra you can emulate the basic stuff you can do with a data-glove, you just lose the fine finger control, but I think it could work.

But, yeah, combining the Hydra and the P5 may work. As bad as the P5 was, the bend sensors did work OK (the positional tracking was horrible though). I would love to just DIY a whole glove, but yeah, only so much time in the day.

_________________
check my blog - cybereality.com


Tue Oct 02, 2012 8:32 pm
Profile
Golden Eyed Wiseman! (or woman!)

Joined: Fri Jul 08, 2011 11:47 pm
Posts: 1498
Reply with quote
Quote:
As bad as the P5 was, the bend sensors did work OK (the positional tracking was horrible though)


That is so absolutely true. Its a shame they didn't fix the tracking before they released it, it could well have done MUCH better if it was actually fit for the purpose. Its only the glove+bend sensors that are useful.

With that Leap tracking thing, it may be easiest to just use that once its released. Hopefully would give you the ability to do pretty much everything that a physical glove does with less setup hassles.


Tue Oct 02, 2012 11:34 pm
Profile
One Eyed Hopeful

Joined: Wed Sep 12, 2012 10:22 am
Posts: 8
Reply with quote
With Leap, i wonder if it would be doable to identify specific objects + hands.
Like sword handle or a gun, custom for a specific game. That could be a nice anti piracy thing or perhaps a addon package, rift compability + whatever you need in the game.


Wed Oct 03, 2012 1:32 am
Profile
Sharp Eyed Eagle!
User avatar

Joined: Tue Aug 21, 2012 6:51 am
Posts: 401
Reply with quote
WiredEarp wrote:
With that Leap tracking thing, it may be easiest to just use that once its released. Hopefully would give you the ability to do pretty much everything that a physical glove does with less setup hassles.

I wonder what the range is.
As far as I know, it's not obvious which technique is used (magnetic, infrared or sonar?)
In all demos I've seen, the hand is within ~30cm range, so it could be a limitation for mobile VR


Wed Oct 03, 2012 4:40 am
Profile
Golden Eyed Wiseman! (or woman!)

Joined: Fri Jul 08, 2011 11:47 pm
Posts: 1498
Reply with quote
Thinking about it, I'm starting to wonder if the Leap will be any good at tracking certain hand positions. For example, if the Leap is on a desk, then if you (for example) had a natural, thumb up hand position, it would probably be unable to determine the thumb position due to it being occluded by the rest of the hand. This may well cause problems with some gestures etc...

I guess we will have to wait for it to be released to determine how effective it will be...


Wed Oct 03, 2012 5:00 am
Profile
Sharp Eyed Eagle!
User avatar

Joined: Tue Aug 21, 2012 6:51 am
Posts: 401
Reply with quote
WiredEarp wrote:
Thinking about it, I'm starting to wonder if the Leap will be any good at tracking certain hand positions. For example, if the Leap is on a desk, then if you (for example) had a natural, thumb up hand position, it would probably be unable to determine the thumb position due to it being occluded by the rest of the hand. This may well cause problems with some gestures etc...


I don't think you have to worry about that too much, as you can see that it is able to determine the shape of your hand pretty well

Image

Quote:
I guess we will have to wait for it to be released to determine how effective it will be...


As usual. Same goes for the Rift ;)


Wed Oct 03, 2012 6:57 am
Profile
Golden Eyed Wiseman! (or woman!)

Joined: Fri Jul 08, 2011 11:47 pm
Posts: 1498
Reply with quote
@ mahler: actually, that doesn't prove anything. The hand position showing in that picture is easy to capture. The one I was mentioning would be much harder, due to the hand occluding the thumb.

For example, giving a 'thumbs up' gesture, while the Leap is mounted below, I dont see how it could read your thumb accurately.


Wed Oct 03, 2012 5:16 pm
Profile
Sharp Eyed Eagle!
User avatar

Joined: Tue Aug 21, 2012 6:51 am
Posts: 401
Reply with quote
WiredEarp wrote:
@ mahler: actually, that doesn't prove anything. The hand position showing in that picture is easy to capture. The one I was mentioning would be much harder, due to the hand occluding the thumb.

For example, giving a 'thumbs up' gesture, while the Leap is mounted below, I dont see how it could read your thumb accurately.
No it doesn't prove anything. As with you probably, I have to see it for myself before I believe it ;)
But if we analyze it, I think the picture shows that even here, when the Leap is on the desk almost below, the top of the hand still seems rather solid. Could be just the software program rendering a model based on key input, but in the best case scenario it just has a type of sensor which just sees a lot of detail. If you look closely, he's not holding his hands right on top over it. So your example could just be solved by not hovering exactly above the device, but rather move it a little to the side. We'll see Jan/Feb next year.


Wed Oct 03, 2012 6:22 pm
Profile
Certif-Eyable!
User avatar

Joined: Sat Dec 22, 2007 3:38 am
Posts: 990
Reply with quote
What's interesting about that picture is that is looks like it's showing the tops of the fungers as well as the bottom. So unless the sensor is in front of the finger tits, I wonder how this is possible, assuming that they are using similar light point tech like the kinect does. It may just be a render for demonstration though, which would be my guess.


Wed Oct 03, 2012 6:27 pm
Profile
Binocular Vision CONFIRMED!
User avatar

Joined: Wed Aug 15, 2012 1:38 pm
Posts: 322
Reply with quote
Or they could just assume basic geometry of a hand. Like a model that's simply controlled through the data input.


Wed Oct 03, 2012 7:13 pm
Profile WWW
Cross Eyed!

Joined: Tue Sep 25, 2012 7:53 am
Posts: 129
Reply with quote
So I scoured the web last night and....not one video of the result when an object is occluded from the sensor. In fact, the people doing the demoing seem to be taking great pains to ensure it doesn't happen, keeping their hands flat etc when more natural movements would lead to obstruction.


Wed Oct 03, 2012 7:18 pm
Profile
One Eyed Hopeful

Joined: Wed Sep 12, 2012 10:22 am
Posts: 8
Reply with quote
Cool!
http://www.youtube.com/watch?v=UmCcYfVXZ5w&feature=player_embedded


Wed Oct 03, 2012 7:58 pm
Profile
Petrif-Eyed
User avatar

Joined: Sat Sep 17, 2011 9:23 pm
Posts: 2220
Location: Menlo Park, CA
Reply with quote


That is pretty nice and a lot more practical for VR than the LEAP. I hope the LEAP team adapts their algorithms to larger devices as well.


Wed Oct 03, 2012 8:02 pm
Profile
Cross Eyed!
User avatar

Joined: Thu Aug 16, 2012 5:30 am
Posts: 101
Reply with quote
I have an idea for a hand tracking method that involves injecting lethal amounts of mercury into the players hands or perhaps a kind of ferrous liquid, that combined with a magnetic tracker.


Thu Oct 04, 2012 6:15 am
Profile
One Eyed Hopeful

Joined: Thu Jul 26, 2012 6:47 am
Posts: 49
Reply with quote
Nogard wrote:
I have an idea for a hand tracking method that involves injecting lethal amounts of mercury into the players hands or perhaps a kind of ferrous liquid, that combined with a magnetic tracker.


Lol, I hear this guy might be able to help you with that. :D

_________________
"Creating the world's most immersive virtual reality erotic encounters @ http://www.wickedparadise.com"


Thu Oct 04, 2012 7:47 am
Profile
Cross Eyed!
User avatar

Joined: Thu May 10, 2012 4:42 pm
Posts: 140
Reply with quote
Nogard wrote:
I have an idea for a hand tracking method that involves injecting lethal amounts of mercury into the players hands or perhaps a kind of ferrous liquid, that combined with a magnetic tracker.

I don't know, that sounds a little dangerous. Instead, how about injecting Technetium 99 in your bloodstream, getting a Geiger counter and bingo! Full skeletal tracking!
Image

In all seriousness, though: has anyone considered using a string system like the Gametrack (does anyone remember it?) mounted on a dataglove for finger tracking? The tracking wouldn't be perfect, but better than the linear reading of a flex sensor and it would be dirt cheap (all you need are potentiometers from analog sticks and strings).

Start from 00:50

_________________
Image
You can also Greenlight other Rift games.


Thu Oct 04, 2012 12:51 pm
Profile
Diamond Eyed Freakazoid!
User avatar

Joined: Mon Sep 03, 2012 1:36 pm
Posts: 741
Location: QC, Canada
Reply with quote
Omarzuqo wrote:
I don't know, that sounds a little dangerous. Instead, how about injecting Technetium 99 in your bloodstream, getting a Geiger counter and bingo! Full skeletal tracking!

:shock:

Kickstart that!


Thu Oct 04, 2012 1:22 pm
Profile
Cross Eyed!
User avatar

Joined: Thu May 10, 2012 4:42 pm
Posts: 140
Reply with quote
MrGreen wrote:
Omarzuqo wrote:
I don't know, that sounds a little dangerous. Instead, how about injecting Technetium 99 in your bloodstream, getting a Geiger counter and bingo! Full skeletal tracking!

:shock:

Kickstart that!

I will. I have a feeling that it will surpass the Rift and the Ouya. ;)

DISCLAIMER:
It can AND WILL produce radiation poisoning, sterility, cancer, genetic mutations within six generations, and death among other side effects if used repeatedly.

_________________
Image
You can also Greenlight other Rift games.


Thu Oct 04, 2012 2:28 pm
Profile
Cross Eyed!
User avatar

Joined: Thu Aug 30, 2012 10:19 am
Posts: 136
Reply with quote
I'm becoming interested in data gloves which use electroactive polymers for haptic feedback.

When you think about how we use our hands in real life, we almost never focus on them. We use our hands like an extension of our vision. When you walk down a dark hall, you can build a picture of it in your mind by feeling along the walls. You can operate precision tools and balance objects without looking at them.

Being forced to watch your hands to see how they're interacting with an object seems like it would be clumsy and slow - I figure the feeling would be similar to using tools while in a Space suit.

This is one of the reasons I can't stand touch screens. When typing on a physical keyboard it's one step to go from your sense of touch to your brain. When you use a touch screen you need to visually locate the key, touch and hold, visually confirm you've pressed the correct key, then release. You can't type fast because you have to visually confirm each press.

Tracking the location of the hands is just a smaller part of the goal of giving the hands sensory information in a virtual environment.

I wonder, if you have decent haptic feedback, will you still be pushing your hand through "solid" walls? Or does your brain adjust and learn to use a light touch to not break through things?


Fri Oct 05, 2012 8:34 pm
Profile
Cross Eyed!

Joined: Tue Sep 25, 2012 7:53 am
Posts: 129
Reply with quote
rhinosix wrote:
I wonder, if you have decent haptic feedback, will you still be pushing your hand through "solid" walls? Or does your brain adjust and learn to use a light touch to not break through things?


Anyone else ever go to move the mouse on their computer when someone else has bumped down the sensitivity and felt like the mouse hand was physically linked to something in your brain...and that something was suddenly encased in mud?


Sat Oct 06, 2012 8:26 am
Profile
Display posts from previous:  Sort by  
Reply to topic   [ 37 posts ] 

Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB® Forum Software © phpBB Group
Designed by STSoftware.