search

autonomousvision / occupancy_networks

This repository contains the code for the paper "Occupancy Networks - Learning 3D Reconstruction in Function Space"
https://avg.is.tuebingen.mpg.de/publications/occupancy-networks
MIT License
1.52k stars 292 forks source link

eval #17

Closed xiayizhan2017 closed 5 years ago

xiayizhan2017 commented 5 years ago

hello, when I run the eval_meshes.py, it has same warning. 'warning: contains1 != contains2 for same points.' I wonder if this warning will affect the results.

xiayizhan2017 commented 5 years ago

Thanks you in advance!

LMescheder commented 5 years ago

Hi @xiayizhan2017, our script for checking if a point is inside or outside the mesh checks the number of intersections with rays going in two different directions. If those two checks lead to different results, then the warning occurs. This either happens because the mesh is not watertight or if there are numerical instabilities. We tried to avoid the later case by writing our code in a numerically stable fashion, but it's difficult to completely resolve it. Nonetheless, these instabilities should only happen rarely and should not affect the overall results.

xiayizhan2017 commented 5 years ago

Thanks!