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

the MWIS part will take a really long time when the number of branches reaches 35 or more. #4

Closed
HuMengjing opened this issue Sep 23, 2019 · 8 comments

Comments

@HuMengjing
Copy link

I run this code with only 5 detections a frame and the N-scan number is 3, and it takes a really long time at frame 3, in which it keeps running the bron_kerbosch() function.
I am not sure if there is a better way for this part or is that because I am not running this codes properly?
And how many detections could the algorithm handle at most when you test?
Really appreciate for your reply.

@jonperdomo
Copy link
Owner

Can you please try lowering the gating area parameter in params.txt?

@jonperdomo
Copy link
Owner

I will test that and get back to you.

@HuMengjing
Copy link
Author

Yeah, actually I set the gating area to 6, and my input is ground thruth(no false alarm), which means there is only one measurement can be included in the gating area.
I'm thinking maybe there is too many vertices and edges for the MWIS( when the number of tracks reach 70, it feels like to loop forever in the bron_kerbosch() function), so I am wondering how many detections a frame could it handle and what is the N-scan window size you set? Thanks a lot :>

@jonperdomo
Copy link
Owner

Yes, I think I will try the NetworkX maximum-weighted matching implementation to see if there is an improvement in performance. For track pruning, I set an N-scan window of 1. I will also add a user-defined parameter for this, but you can modify it here in the code:

K = 1 # Frame look-back for track pruning

@MarwanMansour
Copy link

Any updates on this? I run into the same problem. The MWIS part is taking very long in the second frame with 14 detections.

@asafroth91
Copy link

Any update on this? Its super slow

@jonperdomo
Copy link
Owner

jonperdomo commented Sep 24, 2023

I have added two missing pruning parameters from the paper: Bth for setting the maximum number of branches, and Nmiss for the maximum number of consecutive frames with missing observations for a track in commit 5627c53. Please try adjusting these parameters to improve performance.

@HuMengjing
Copy link
Author

HuMengjing commented Sep 24, 2023 via email

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