Giters

dazinovic / neural-rgbd-surface-reconstruction

Official implementation of the CVPR 2022 Paper "Neural RGB-D Surface Reconstruction"

Home Page:https://dazinovic.github.io/neural-rgbd-surface-reconstruction/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Evaluation steps

Yiiii19 opened this issue · comments

commented

Hi,
firstly thanks so much for this amazing work.

Now I finished all the preparation steps and started running "optimize.py".

  1. After traininng I will get a reconstruction mesh which is exactly as same as the "meshes.zip"?
  2. If not, which step should i do to get the same result from the zip file?

Thanks for your reply.

commented

Since the optimization process is always a bit noisy, you won't get exactly the same result, but it should be very similar. One possible difference is the coordinate system (you might have to rotate the mesh around some axis (probably x-axis by 90° or -90°) or scale and translate it using the parameters from the config file).