nmwsharp / learned-triangulation Goto Github PK

Could you provide your DataSet when you training model used and the test.ply?

Thanks!

Hi! Thank you for your great work! Your demo gif is so cool! May I ask how can you make that? Can you share the visualization script?

How could I measure the watertight rate of result mesh ?

Dear Nicholas and Maks,

I would like to thank you for your outstanding work. It's a very valuable contribution to the computer vision community, and thanks for making the code open source.

i) In my study, I will use the same comparative baseline as in the paper (Table 1 page 11). It is unfortunate that I cannot locate the corresponding code for these three geometric methods: ballpivot, alpha-3, and alpha-5.

Maybe I did not check the comparison baseline properly, if so, could you please provide me with the script?

ii) To estimate the BAP radius value. Could you please provide references for the scheme you have used [For ball-pivoting, the ball radius is automatically guessed as the bounding box diagonal divided by the square root of the vertex count] Or was it selected based on general understanding?

I really appreciate your help!

Houda

Hi,

First of all I would like to congratulate you on the paper! I think this contribution is great. I was planning to use this implementation strictly for research purposes (eg. comparing results and obviously cite your paper) but there's no license attached to the repo. Would you mind adding one?

Thanks!

Can you write a detail build description?

I can't work when I follow the README.

Thanks !

The following packages were not installed (Ubuntu 20.04LTS, conda 4.9.2):
I wasn't able to install them by adding the channels in the environment.yml, but this fixed it:

conda install -c Esri torch-cluster torch-scatter
conda install -c pytorch pytorch pytorchvision