cleinc / bts

From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

About the LPG

YiLiM1 opened this issue · comments

Thank you for your great work, but in the LPG module, the normal vector you generate is in the camera coordinate system, while u, v seems to be in the pixel coordinate system. How do you connect the two? I don't seem to see you use camera parameters.

I'm curious to this issue too.

I'm curious to this issue too.

Got the same confusion. Equ 1 seems to be correct only in camera coordinate system, otherwise n1~n4 do not represent plane coefficients

same confusion, how to get n1~n4 without camera parameters?