search

autonomousvision / convolutional_occupancy_networks

[ECCV'20] Convolutional Occupancy Networks
https://pengsongyou.github.io/conv_onet
MIT License
830 stars 113 forks source link

Post-Processing #44

Closed ceteke closed 1 year ago

ceteke commented 2 years ago

Dear authors,

Thank you for this nice work and sharing the code with the community! I wonder if there are any post-processing steps done (e.g. smoothing the meshes) after the 3D reconstruction, which might not be mentioned in the paper.

Best regards

pengsongyou commented 2 years ago

Hi @ceteke

Thanks for your interest. There is no post-processing operations after the marching cubes. We indeed had a refine_mesh function (copy from the original Occupancy Networks repo):

https://github.com/autonomousvision/convolutional_occupancy_networks/blob/f15b97da2c5995598537c8f832e52e95c0b09236/src/conv_onet/generation.py#L419-L423

But we always set the refine to false, e.g. check this config.

Best, Songyou