Skip to content
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.

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

There is a lot of noise around the object reconstructed by neus-acc. #296

Open
wujinhu1999 opened this issue Mar 7, 2024 · 5 comments
Open

Comments

@wujinhu1999
Copy link

Hello authors, thank you for your great work! I encountered some problems while training neus-acc. There is a lot of noise around the reconstructed object. How to remove this? Are there any hyperparameters that need to be adjusted? Looking forward to your reply. My training command is as follows:ns-train neus-acc --pipeline.model.sdf-field.inside-outside False --vis wandb --experiment-name neus-acc-dtu24 sdfstudio-data --data data/dtu/scan24
问题

@niujinshuchong
Copy link
Member

Hi, these regions correspond to the white table and black background in the input images. Therefore, no correspondence can be found and the optimised surface is kind of random. You can remove it with object masks.

@wujinhu1999
Copy link
Author

Hi, these regions correspond to the white table and black background in the input images. Therefore, no correspondence can be found and the optimised surface is kind of random. You can remove it with object masks.

Hello, it means use the command '--include_foreground_mask True'?

@hanjoonwon
Copy link

hanjoonwon commented Mar 11, 2024

Hi, these regions correspond to the white table and black background in the input images. Therefore, no correspondence can be found and the optimised surface is kind of random. You can remove it with object masks.

Hello, it means use the command '--include_foreground_mask True'?

Can you tell me the train option?
i trained with ns-train neus-facto --pipeline.datamanager.train-num-rays-per-batch 2048 --pipeline.model.sdf-field.use-grid-feature False --pipeline.model.sdf-field.inside-outside False --pipeline.model.background-model mlp --pipeline.model.mono-depth-loss-mult 0.0 --pipeline.model.mono-normal-loss-mult 0.01 --vis wandb --trainer.steps_per_save 5000 --trainer.steps-per-eval-image 5000
but I'm getting very strange results.

@wujinhu1999
Copy link
Author

For the neus-facto model, you can just use the author's default commands. My result comes from the neus-acc model.

@hanjoonwon
Copy link

For the neus-facto model, you can just use the author's default commands. My result comes from the neus-acc model.

Thank you i don't know why my outputs are so poor

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants