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.

Sign up for GitHub

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

Jump to bottom

Temporal inference #12

Closed
BaoWentz opened this issue Apr 3, 2023 · 5 comments
Closed

Temporal inference #12

BaoWentz opened this issue Apr 3, 2023 · 5 comments

Comments

@BaoWentz
Copy link

BaoWentz commented Apr 3, 2023

Can we futher improve mAP or NDS of valset and testset, if we use temporal information during inference?Thanks!
@linxuewu
Copy link
Owner

linxuewu commented Apr 3, 2023

Which specific temporal information are you referring to?

@BaoWentz
Copy link
Author

BaoWentz commented Apr 3, 2023

Sparse4D/projects/configs/sparse4d_r101_H4.py

Line 24 in c48ecd2

seq_frame=max_queue_length,

During test time, we set seq_frame=0 by default, so we won't use temporal keypoint information.

@linxuewu
Copy link
Owner

linxuewu commented Apr 3, 2023

During inference, instead of using multi-frame sampling, we employed sequential reasoning and stored the features of historical frames in the head. Therefore, we utilized temporal information in our approach.

@linxuewu
Copy link
Owner

linxuewu commented Apr 3, 2023

https://github.com/linxuewu/Sparse4D/blob/main/projects/mmdet3d_plugin/models/sparse4d_head.py#L171

@BaoWentz
Copy link
Author

BaoWentz commented Apr 3, 2023

I got it, thanks for your reply.

@BaoWentz BaoWentz closed this as completed Apr 3, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@BaoWentz @linxuewu