What data i will need to get real world cordinates of object in a new image ?

[arun121cs]

No description provided.

[GSORF]

Hi,

using two view reconstruction it is only possible to reconstruct the 3D points up to an arbitrary / unknown scale.

In order to do that you need a calibrated camera (otherwise you would get a projective transformation instead of a euclidean / metric), corresponding points in both images and then you can use the code in https://github.com/alyssaq/3Dreconstruction/blob/master/example.py directly to triangulate your points.

@alyssaq : Thank you for providing your source code. Btw, what is the license?

[alyssaq]

MIT 😄