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 Parameters #1

Closed
HyunsooCha opened this issue Apr 29, 2022 · 2 comments
Closed

About Parameters #1

HyunsooCha opened this issue Apr 29, 2022 · 2 comments

Comments

@HyunsooCha
Copy link

HyunsooCha commented Apr 29, 2022

Hi, thank you for your excellent work!

I have some questions related to parameters.

  1. In a rigid pose, the rotation parameter's dimension is the torch.Size([1, 3]). What is the type of this rotation? I was confused that the type is radian angle (yaw, pitch, roll) or angle-axis.
  2. In a translation, what is the unit of this parameter? Pixel or cm?
  3. The camera intrinsic parameter yielded [0.31053847 0.50841707 0.5909382]. I guess the first one is focal length, and the other is cx, cy. However, when I used the MATLAB toolbox by CalTech, I got the focal length to 2463 as the same dataset. What is the type or unit of focal length on your code?

I'm so happy to know your excellent work! This code is really helpful to me.

Thank you.

@philgras
Copy link
Owner

philgras commented May 2, 2022

Hi!

  1. Its axis-angle representation. The batch_rodrigues function in vht/model/lbs.py gets you the rotation matrix.
  2. The translation vector as well as vertex coordinates of FLAME are given in meters.
  3. [f, cx, cy] is right. The values are normalized. To denormalize them for a resolution of [h,w], do f * max(h, w) and cx * w and cy * h. These values are then in pixel units.

Hope this helps.

@HyunsooCha
Copy link
Author

Thank you for your reply! It is beneficial to use the code.

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