autonomousvision / giraffe

This repository contains the code for the CVPR 2021 paper "GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fields"
https://m-niemeyer.github.io/project-pages/giraffe/index.html
MIT License
1.23k stars 160 forks source link

config difference between giraffe and graf on celebahq #28

Closed MrTornado24 closed 3 years ago

MrTornado24 commented 3 years ago

Hi, thanks for sharing the great work. I wonder why there are some difference of config parameters on celebahq, e.g. the near, far bounds, range of u, v. How you determine the min max bounds of camrea pose and object2scene transformation of celebahq dataset in giraffe?

m-niemeyer commented 3 years ago

Hi @MrTornado24 , thanks for your interest in our project! We mostly changed things to be more congruent with the other datasets, e.g. the elevation angle of 10 degrees we use across many datasets. Presumably changing these things slightly doesn't change results much!