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

about feature extraction of THUMOS14 #7

Open
menghuaa opened this issue Mar 24, 2022 · 4 comments
Open

about feature extraction of THUMOS14 #7

menghuaa opened this issue Mar 24, 2022 · 4 comments

Comments

@menghuaa
Copy link

Excuse me, what methuod do you use to extract the feature of thumos14? Is it I3D?

@Pilhyeon
Copy link
Owner

Yes, we used the two-stream I3D networks pre-trained on Kinetics-400, as stated in the paper.

@menghuaa
Copy link
Author

Yes, we used the two-stream I3D networks pre-trained on Kinetics-400, as stated in the paper.

ok,thank you.You said in your paper that you take 16 frames of video as a segment, and then extract features. But when I output the shape (nxdimension) of the feature, n is not equal to the total number of frames /16. Why is this?

@Pilhyeon
Copy link
Owner

It is probably because we set the fps of all videos in THUMOS'14 to 25 before extracting the features.

@menghuaa
Copy link
Author

Yes,that makes sense!Thank you very much!

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