Error maps calculation

[sbharadwajj]

Hi,

I had a few questions about how the error map is calculated:
from here it is understood that the formula to calculate error maps is:
D_dash - D_projected / D_projected + 10e-7, and from your paper, D_dash refers to the depth priors obtained from the monocular depth network that is indexed with the 2D pixel coordinates, and D_projected refers to the depth maps obtained from projecting from eg: view-0 to view-1,2,3,4..etc

I hope this is correct? because my questions are based on the assumption that the above is what is implied by the paper code, if so I will continue with my questions :)

[weiyithu]

Yeah, it is correct.