I'm looking for a way to get 3D coordinates from a 2D image. Because I need to find the normal vectors of several surfaces in the 2d image (if there is a way to find normal vectors without 3D coordinates, I'd love to hear about it).
The conditions are as follows: The 2D image is from a regular camera. Only one frame is available. No depth information is available. Camera calibration is available. You are using Python (but if you have a way to use another language, please let me know so I can use it as a reference).
Is it possible to get 3D coordinates from a 2D image under the above conditions?
My searching skills don't allow me to find what I'm looking for, all the methods I found require depth information. So if you know how, please let me know.
I haven't written and run the code, but I've thought about using a depth estimation model(DL)... But I thought it would be too complicated, so I came to ask if there is a simpler way to do it.