Absolute depth conversion

[alexmagsam]

Hello,

Can you please elaborate on the conversion to absolute depth for the use case of benchmarking on other datasets? I am unsure what the output represents other than the explanation of a 'inverse relative depth'.

Thank you, Alex

[ranftlr]

Have a look at the discussions in #36, #37 and #42.

In the context of benchmarking you could do what we did in the paper: estimate a scale and shift for every image that aligns the prediction to the ground truth under some criterion before measuring errors. See my last response in #36.