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

Reproduce Co-Fusion paper result on car4 and room4 #25

Open
ung264 opened this issue Nov 9, 2023 · 2 comments
Open

Reproduce Co-Fusion paper result on car4 and room4 #25

ung264 opened this issue Nov 9, 2023 · 2 comments

Comments

@ung264
Copy link

ung264 commented Nov 9, 2023

Hi,

I managed to run and build the software in ubuntu 20.04. However, when I evaluate the results for car4 and room4, I found that the camera AT RMSE is 5cm and 14cm, which are much higher than the error indicated in the paper. I used scripts from https://github.com/martinruenz/dataset-tools to evaluate the result for RGBD image sequence of car4 and room4.

I have tried both the default command line arguments and the example command line arguments from dataset tools by adding
-segMinNew 0.008
-crfSmooth 1
-crfPos 0.51
But neither of them gives same results as mentioned in the paper.

Do you have any idea what could be the reason? Could you share the parameters that I can use to reproduce the result? Thanks!

@EdwardjkFeng
Copy link

EdwardjkFeng commented Dec 1, 2023

Hi,

I also tried to run it on these two sequences recently. I used the example command line arguments from dataset tools as well. The ATE of car4 and room4 I obtained are 2.7cm and 1.4cm respectively. Although they are still not as good as the ones reported in the paper, they look more promising than yours.

Here's the arguments I used:

./CoFusion -basedir path/to/car4-full/ -dir colour -depthdir depth_noise  -run -ep -colorprefix Color -depthprefix Depth -cal calibration.txt -segMinNew 0.008   -crfSmooth 1   -crfPos 0.51 -exportdir estimate -maskprefix nomask -es -keep

Could you share more details about the command you used to run it? For example, did you specify the calibration parameters using -cal. This would be a possible reason you obtained such a large ATE.

@ung264
Copy link
Author

ung264 commented Jul 9, 2024

Hi,

@EdwardjkFeng Thanks for your reply. I was on another project and just started to pick up this project recently.

I am using exactly the same command as you. I build the project in docker nvidia/cuda:11.5.2-devel-ubuntu20.04 image.
Here is the dockerfile
Dockerfile.txt

Here are the estimated pose files.
poses-0.txt
poses-1.txt
poses-2.txt
poses-3.txt

Could you maybe share more information about your environment, and your estimated pose files? Then I can evaluate the error on your estimated pose file at my side, to rule out any problem in the evaluation script.

Looking forward to your reply. Thanks!

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

2 participants