Patrick Labatut

Directly using public datasets for neural rendering is something we are actually looking into.

Only a single texture is supported at this time. One could however work around this by composing together the makeup + face texture and apply the resulting texture using PyTorch3D.

Thanks for the suggestion and for sharing some information on your use case! Environment lighting is definitely on our radar as a potential future feature. More specifically, diffuse irradiance implemented...

> [...] I am not sure the results are correct or not and not know how to verify them Ensuring the computations are correct with documented conventions to facilitate interoperability...

> Further, the scale of the coefficients might not have a standard AFAICR a factor of 1 / π may or may not be pre-applied to irradiance maps / irradiance...

> By the way, does there already have any plans to support SH light? Yes, as mentioned in the above comment, this is definitely on our radar for future releases.

@bottler If there are other similar reports (looks to me this is the first one), could this be caught earlier / with a more explicit message?

That makes sense. A simple way to achieve this would be to rename the existing `chamfer_distance()` function into e.g. `chamfer_distance_xy()` and change its implementation and return type to be a...

> Why would you want to compute the F-score on the cham_x and cham_y separately? My understanding is that @gaetan-landreau would like to get access to **both** `cham_x` and `cham_y`...

Closing as answered.