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

Some questions #18

Closed
limjaebin opened this issue Feb 17, 2022 · 1 comment
Closed

Some questions #18

limjaebin opened this issue Feb 17, 2022 · 1 comment

Comments

@limjaebin
Copy link

limjaebin commented Feb 17, 2022

hi, thank you for publishing this paper.

As I read this paper, I have a question.

  1. Can I know the model used in Table 1 and the model used to measure Full in Table 3 ?

  2. Can I know what hit means in Table 1 ?

Thank you for answering

@Flowerfan
Copy link
Owner

  1. The model trained with human single frame annotations in Table 1 is the model in the last low of Table 3. You can get the model by running python main.py. The model is a bit better than the model in the original paper.
  2. Hit is the new metric we proposed for single frame evaluation. For each detected action instance, we select the single frame of highest actionness score for testing. Hit means the selected action frame is in the time range of groundtruth. You can also check the code for more details.

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