nv-tlabs / NKSR

[CVPR 2023 Highlight] Neural Kernel Surface Reconstruction
https://research.nvidia.com/labs/toronto-ai/NKSR
Other
765 stars 44 forks source link

Question about the usage of pycg in this project #13

Closed JIANG-CX closed 1 year ago

JIANG-CX commented 1 year ago

Wonderful work! "I have some questions about the "pycg" package that is used in this project. Are there any special properties that differentiate it from the original "open3d" package? Also, is there any usage guidance available for "pycg"? I have only found this link: https://pycg.readthedocs.io/en/latest/, but it does not provide much guidance." Thanks

heiwang1997 commented 1 year ago

Hey thanks for your interest in our work!

Basically pycg is a utility package, and pycg.vis is a wrapper of Open3D that simplifies its API for fast visualization and rendering. The customized version (source code available at https://github.com/heiwang1997/Open3D) of Open3D differentiates from the official version in that I've incorporated many changes that supports multi-window camera synchronization, animation, scalar analyzer, ... I think you can use the official Open3D and it will also work fine!

Sorry I failed to find a good time to write docs for pycg -- I will try to populate the doc next week :)

rohitdhote111 commented 1 year ago

how to save colored mesh using pycg vis?

heiwang1997 commented 1 year ago

https://github.com/nv-tlabs/NKSR/issues/3#issuecomment-1585991146

vis.to_file(vis.mesh(mesh.v, mesh.f, color=mesh.c), "a.ply")

Please note that pycg.vis is not a necessity for visualization. You can keep using your favourite 3d visualization library.

JIANG-CX commented 1 year ago

Thanks for your answer! (^▽^)