search

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
1.5k stars 292 forks source link

High frequency details #19

Closed nicolasugrinovic closed 4 years ago

nicolasugrinovic commented 5 years ago

Hi, thanks for the great work! I have trained the network generating data with the per-processing script and the end result looks good but it lacks details. Is there something I could/should do to improve the capture of details? Is there a parameter(s) to be changed in the per-processing step? Or something of the like. Can MISE parameters can achieve more details? Thanks!

LMescheder commented 5 years ago

Hi @nicolasugrinovic, your question is hard to answer without additional details about your goals / current status. Did you train on your own data? How do the results currently look like? To get a general sense what details you can expect, take a look at the pretrained models. In general, longer training helps a lot to get more details. Moreover, it can be helpful to increase the number of upsampling steps for inference in your config file.