autonomousvision / occupancy_networks

This repository contains the code for the paper "Occupancy Networks - Learning 3D Reconstruction in Function Space"
https://avg.is.tuebingen.mpg.de/publications/occupancy-networks
MIT License
1.48k stars 291 forks source link

Reconstruction loss for variational inference (PointNet case) #125

Open space1panda opened 1 year ago

space1panda commented 1 year ago

Hello,

Could you please elaborate on the absence of reconstruction loss in the objective function? The BCE loss is between gt occupancy and predicted occupancy, it is not a reconstruction of geometry - so how exactly we guarantee that encoded latents represent pointclouds? Thanks for the answer