search

NVlabs / nvdiffrec

Official code for the CVPR 2022 (oral) paper "Extracting Triangular 3D Models, Materials, and Lighting From Images".
Other
2.09k stars 222 forks source link

Use on MipNerf360 dataset? #133

Open antithing opened 1 year ago

antithing commented 1 year ago

Hi, and thanks for making this code available. Is this approach suitable for larger scenes, such as the mipnerf360 dataset, or outdoor locations? Or is it more focused on object capture?

Thanks!

jmunkberg commented 1 year ago

Thanks for your comments.

This approach targets object capture and will not produce great results on large outdoor scenes out of the box. Things may improve a bit with some background representation, like what is used in mipnerf360, but that it not part of the current code base.

antithing commented 1 year ago

Thank you! One more question, is there any example showing how to use of custom data?

Can I pass the same data as instant ngp?

jmunkberg commented 1 year ago

Hello,

We currently do not provide detailed instructions on how to run with your own images, but it has been discussed in other issues:

https://github.com/NVlabs/nvdiffrec/issues/33 https://github.com/NVlabs/nvdiffrec/issues/31 https://github.com/NVlabs/nvdiffrec/issues/26 https://github.com/NVlabs/nvdiffrec/issues/3