Scale world points properly to compute eucledian distances

[danielsanjosepro]

For some reason the reconstruction has wrong units now. I haven't looked at it in detail. Just adding this issue to do it soon.

[danielsanjosepro]

Not solvable: if the images are taken with a single calibrated camera, then the 3-D structure and camera motion can only be recovered up to scale. To get the scale we need to detect an object with a known size and scale the rest. This makes measuring distances actually much more complicated. Potential solutions:

• Ask the user to mark an object and estimate its size
• Detect an object (Deep Learning) and estimate size