jonbarron

You are correct! Thanks for reporting this. I think it's fixed now, please let us know if you have any other issues.

We haven't tried, we only use Linux. Feel free to post error messages and maybe some other windows user might have an idea.

1. This just dilates the mask, https://en.wikipedia.org/wiki/Dilation_(morphology). By dilating the mask and then removing the original mask, you get a boundary around the ellipse. 2. `mahal_dist == min(mahal_dist(:)` just makes...

Yes, if you set that flag the code will expect a .txt file with features in it, and will crash if it doesn't exist.

Wow, thanks for identifying this! Definitely a very important bug. Are you able to re-run the fixed code on any of the existing benchmarks (Gehler Shi or Cheng)? Updating the...

Thanks for the heads up! Would you mind pushing a CL with that change to requirements.txt?

Yeah I guess you could directly change that function to have a different shape, but it's used during training so I'm not sure what effect that will have. The easiest...

That's weird. I wouldn't worry about it, the errors here seem small enough that they're probably due to numerical instability of the hardware running the code (maybe the tests are...

What hardware are you running this on? The test breakage in the dataset code looks unusual to me. Do you have multiple GPUs? I think the unit tests might assume...

Turning a nerf into a textured mesh is still an open research question.