How do you create the object masks in the DTU dataset ?

autonomousvision / differentiable_volumetric_rendering

This repository contains the code for the CVPR 2020 paper "Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision"

http://www.cvlibs.net/publications/Niemeyer2020CVPR.pdf

MIT License

794 stars 90 forks source link

How do you create the object masks in the DTU dataset ? #36

Closed phongnhhn92 closed 3 years ago

phongnhhn92 commented 3 years ago

000 Hi, I am interested in your labeled mask of 3 scans from DTU dataset. I would like to ask how did you label it. My task is to get similar mask for the whole dataset (~100 scenes).

m-niemeyer commented 3 years ago

hi @phongnhhn92 , thanks a lot for your interest in our project!

We labelled them ourselves as indicated under 4.2 - Dataset of the main publication, so sadly I can't point you to an automatic process for this. We used the open-source tool Gimp for creating the object masks.

Good luck with your research!