search

nexuslrf / ENVIDR

Implicit Differentiable Render with Neural Environment Lighting
https://nexuslrf.github.io/ENVIDR/
MIT License
77 stars 3 forks source link

About the processed sphere data and pretrained neural renderer #3

Open Beniko95J opened 1 year ago

Beniko95J commented 1 year ago

Hi, thank you for the great work. I failed to run python generate_set.py, given the following error messages:

[Open3D INFO] EGL headless mode enabled.
FEngine (64 bits) created at 0x7f6cb07ef010 (threading is enabled)
eglInitialize failed
Segmentation fault (core dumped)

I am still finding the solution (It would be appreciated if you could help me with this), but may I ask for the processed sphere data and the pre-trained neural renderer model? It seems that no links are provided in the README.md.

Thanks!

nexuslrf commented 1 year ago

Hi, Thanks for your interest! This EGL issue is likely to be related to the OpenGL setting for headless Nvidia GPU. For example, your machine might have installed the Nvidia drivers with flags such as --no-opengl-files. You can reinstall Nvidia drivers with proper opengl support to fix this error.

On the other hand, I'll upload the pre-trained renderer model first this week for further help. :smiley: I will notify here once done.