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 the function <normalize_xy_rotation> and <denormalize_xy_rotation> #12

Open
allenhsu6 opened this issue Nov 18, 2024 · 0 comments

Comments

@allenhsu6
Copy link

Hi @jmwang0117, Thanks for the great work!

I have some questions about the work.

I found that no matter in the training or testing process, the trajectory will first be scaled + rotated, and then the pyramid_noise_like function will be used to generate noise. I would like to ask you the following questions:

  1. Why do you do this? (Is it because after rotating the interval 360 degrees to cover the two-dimensional plane, adding noise will be more in line with the Gaussian distribution?)
  2. Will the trajectory after uniform rotation not lose the physical meaning of its own orientation? For example, the orientation information of the road originally contained in the img feature condition.
  3. Is the trajectory represented by the absolute position or the position relative to the previous point (the interval between the two points before and after)?
  4. How to choose scale x and scale y?
  5. Why are the values ​​of scale x and y different in normalize and denormalize?
    Looking forward to your reply, 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

1 participant