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

bad canonical image? #49

Closed
rainsoulsrx opened this issue Aug 25, 2023 · 3 comments
Closed

bad canonical image? #49

rainsoulsrx opened this issue Aug 25, 2023 · 3 comments

Comments

@rainsoulsrx
Copy link

Hi, I test reconstruction using this video, though reconstruction is ok, but I get strange canonical image as follows

walf_base_dual.mp4

image

@ken-ouyang
Copy link
Collaborator

Please refer to #31.

I have a few additional suggestions that might help enhance your results:

  1. Consider segmenting the objects initially. This approach can help prevent unusual, isolated boundaries from appearing in the canonical images, which could lead to more accurate reconstruction.

  2. For the sequence you're working with, using two canonical images might yield better results. We're currently in the process of integrating 3D-awareness into the structure, which should further improve the outcomes.

  3. Despite these potential improvements, it's worth noting that your current reconstruction seems to be of good quality. Therefore, you might still want to proceed with editing and then observe the translated results.

@pHaeusler
Copy link

@rainsoulsrx - I have some notes that might help https://philliphaeusler.com/posts/codef/
https://github.com/pHaeusler/codef-experiments

@jetterson
Copy link

jetterson commented Dec 11, 2023

@rainsoulsrx Did you try to reconstruct the video use the canonical image you have get? I tried but the performance of the output video is bad, did you meet the same problem?

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

4 participants