Figure 2
The pipeline of sparse GUSL, a multi-grid representation learning process. Note that the Saab Hop module and the 2x2x2 pooling/upsampling operations are based on sparse voxel operations, where empty voxels are not processed. Refer to the image caption for details.

The pipeline of sparse GUSL, a multi-grid representation learning process. Note that the Saab Hop module and the 2x2x2 pooling/upsampling operations are based on sparse voxel operations, where empty voxels are not processed.

or Create an Account

Close Modal
Close Modal