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

Foreground and Background meshes are empty #186

Open
Heba2h opened this issue Apr 12, 2024 · 3 comments
Open

Foreground and Background meshes are empty #186

Heba2h opened this issue Apr 12, 2024 · 3 comments

Comments

@Heba2h
Copy link

Heba2h commented Apr 12, 2024

Screenshot from 2024-03-12 16-52-27
To begin, I processed the video using Colmap with the following commands:

Convert the video to individual frames:

ffmpeg -i /project/gaussian-splatting/Museum.mp4 -vf fps=2 -vsync vfr -q:v 2 /project/gaussian-splatting/data/input/%06d.jpg

Convert frames to a suitable format:

python convert.py -s /project/gaussian-splatting/data

Next, I utilized Gaussian Splatting to generate the Gaussian representation of the scene:

Perform Gaussian Splatting:

python SuGaR/gaussian_splatting/train.py -s data/ --iterations 5000 -m gaussian_output/
Then, I executed SuGaR:

Run SuGaR with specified parameters:

python SuGaR/train.py -s data/ -c gaussian_output/ -r "density"

Please review this workflow and identify any potential issues that make the background and foreground meshes empty.

@CDchenlin
Copy link

Same problems

@tianf-code
Copy link

Same problem. Did you solve it?

@Heba2h
Copy link
Author

Heba2h commented May 13, 2024

@tianf-code
Yes we solved it, and mesh extraction was done successfully.
The problem was from the output of colmap, it showed us that it was successfully done but actually the output from comlap alot of things were missed.
U can check undistorted images if there’s many images or not

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