thank you for the app. I have not tried it yet. As a neurolgist, I was ondering if this could be made into an inexpensive tool for people with cervical spinal cord lesions. There are systems that exist but most are expensive. I look forward to your thoughts and comments.

where is the functions llist? and manual do you have something? so i can know how to i get the image to panel for example? if thats possibble at all.?

the points are merget into the same pixel center, but when you move appart, they will reform again, i hope.

I've tried to track multiple (two) points, when trying to get the distance of the two spots to the camera.

Since the points will swap if one gets brighter then the other, I couldn't get this to work as it should. That happens a lot if you use two torches or two white spots on a paper, depending on the angle to the camera.

So if two spots could be identified in an absolute way, so that it always knows which point it is, it would help a lot for mocap or any other precise tracking.

I think there is a function that determents the points color, but since that can change with the points brightness, I don't think it would solve the problem.

well you can always analise the 4d dimension right? the time...

So, you can run/apply a probability prediction of who is who. istead of random, at least you could get more accurate result.

So you know that if you mix 2 points one coming from above to down, and one from the left to the right..and they cross to a single point.. the next value from that cross, is what? probably the dots would keep their direction, so the above will keep goin down, and the left will keep going right, and not, the up will go right and the left to right will go down.. understand me?

So you have the var delta_x and delta_y that holds the position frame -1 frames, so you get a direction vector, and you then predict based on that. like(if the direction vector is the same then id is the same, else switch).

I think your idea could work if it's done right and if those very unlikely situations of points changing direction just when they're crossing, are not messing things up.
That combined with a color recognition should get results that are reliable enough

Since in my case I'm only using points that will never cross their path, I could fix my problem by always taking the right point as point one, instead of the bright one.
This works perfectly, I can know get the distance between the light sources and the camera. So if the observer has two light sources left and right of the head, then I know if he's moving closer or further away.

Now the next problem is that if he's turning the head, then the points are moving closer together and it looks like if he's moving closer.
I have to find a solution for that yet...

Oh yes, that because you are using 2 points, and you cant derivate that, from 2 points..
So your solution is very very simple..

You must have have 3 points.. So.. if you have a head with 2 lights, you need on more in on the upper side.. like this:



()
||
()--[HEAD]--()

Thus you can draw a triangle.. and when you aproach , however they scale they on screen.. the distance between those, is the same...
But when you rotate.. the distance will cause the triangle to srhink.. horizontaly and the distance of the left to the right will be smaller but, the up will mantain..

You can calcultate that by the angles of the corners, or by dividing the side lenghts with the down lenth

I guess that would be a solution.
But it would require a lot of calculations and I would look like Sam Fisher

I'm thinking more of something like the Wii-Bar, which is essentially just like the white spots for the camera. And since I would like to reduce the amount of equipment needed, I will try to get it to work with I guess about 4-6 points.
They could then be mounted on sunglasses, or just white spots on a black paper.

The Wii-Bar contains 10 lights, 5 on each end. I think the way they do it is looking at how much the points a getting closer together. If they all come closer together at the same speed, then they are all getting closer. If the points on one side move closer together faster than on the other, that means this side has moved closer while the other one has moved away, so the player has turned.
Something like that...

But I'm not sure if tracking that many points, with the small resolution of a webcam, is going work out.

white spots on a black paper? why dont you use an infra red reflting paper, that you can easly crop and glue tot he sun glasses, .. and so enchace the IR you can use a tv control to send light in front of you..

My main goal is to keep it as simple as possible. I would like to keep this down to a webcam and at best two torches. But I realize that it doesn't work with the distance, so I have to make it a bit more complex, but I'd rather not touch IR.
Other wise the Wii-Remote solution would solve all my problems.

But I just realized that when the observer moves left and right in front of the screen, he always turns towards the screen and the webcam, if he's directly in front of it, he wont turn he's head away. (or shouldn't)
So maybe I can just account for the amount of head turning when he's moving to the side, that the distance stays steady enough.

It's not perfect, because the angle changes when getting closer to the webcam, but then again, this is not that precise, and it doesn't have to be, just close enough.