search

autonomousvision / convolutional_occupancy_networks

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

Train on custom dataset using crops #41

Open raphaelsulzer opened 2 years ago

raphaelsulzer commented 2 years ago

Hi,

I have successfully trained ConvONet on my own object datasets before with nice results! I have used the sample_mesh.py script from ONet to generate pointcloud.npz and points.npz (occupancy points) files.

I would now like to train the network to reconstruct very large scenes, for which I have ground truth meshes for some of them. How do I go about this?

Can I simply provide one points.npz and pointcloud.npz file per scene? How do I make sure to have enough occupancy samples per crop? Should I simply make sure to have 100k occupancy points per crop defined by voxel_size * resolution? Or do I need to do the cropping myself?

Kind regards