New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Training for RGBA images #89

Closed

hardikdava opened this issue Dec 21, 2023 · 1 comment

Contributor

hardikdava commented Dec 21, 2023 •

edited

Thank you for your amazing work.

Is it possible to train for RGBA images? e.g. with alpha channel.
If not how can it be done?

ivanpuhachov commented May 30, 2024

It's an old issue but I was doing the same recently.

rasterize_gaussians returns accumulated alpha (1 - final_Ts) if you call it with return_alpha=True
see splatfacto training for more examples https://github.com/nerfstudio-project/nerfstudio/blob/5491df92d9f6a47c86f7e1ecb4e986ae1e1f5632/nerfstudio/models/splatfacto.py#L776
see discussions on alpha carving in splatfacto Add transparency carving to splatfacto nerfstudio#2889

I think this issue can be closed.

hardikdava closed this as completed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment