I'm working on a project where I need to find the coordinates of an object.
I used this paper for now http://dsc.ijs.si/files/papers/S101%20Mrovlje.pdf . It describes how with two cameras facing the same way, you can get the distance from the object to the axis they're aligned on.
From there, I wondered if it was possible keeping the same hypothesis to get a second coordinate so that I know the distance between the projection on the axis and a camera. Here's what I came up with : https://i.stack.imgur.com/PkxN1.jpg
Now, if this works, I can get a second distance. What I am unsure about in what I wrote is the line that uses the number of pixels to determine a length (assuming the camera doesn't have any distortion). Would this be true ?
If so, the rest of the computation should be pretty straightforward, and I hope good.
Now, the other question I have is : what if we don't consider the hypothesis to be true, and there is an angle between the two cameras (That we would obviously know). The relationship in the paper that gives us the distance L would be wrong. Is there another relationship that can be found to compute said distance ?
Last question : If a third camera is added, not in the same plane, is it enough to compute the missing coordinate ?
Thanks for your time