video-to-pose3D icon indicating copy to clipboard operation
video-to-pose3D copied to clipboard

Published 20 hours ago • verified publisher icon zh-plus

Normalize/unNormalize Data

Open mmaGroupGitHub opened this issue 3 years ago • 1 comments

Could you explain and point to me in your code where you are perform normalization and unNormalization of the data.

If I just try to plot the 3D output the values are all less than 1 just trying to understand where you perform unNormalization and also what sort of preprocessing you apply to the two dimensional keypoints.

mmaGroupGitHub avatar May 28 '21 17:05 mmaGroupGitHub

The 2D normalization: https://github.com/zh-plus/video-to-pose3D/blob/e271582b762f9f56b50f75f8d44cc351736d8683/videopose.py#L79

The 3D unNormalization may be performed here: https://github.com/zh-plus/video-to-pose3D/blob/e271582b762f9f56b50f75f8d44cc351736d8683/common/utils.py#L191

The 3D Normalization in the function model_pos: https://github.com/zh-plus/video-to-pose3D/blob/e271582b762f9f56b50f75f8d44cc351736d8683/common/utils.py#L189

I'm not sure what "Normalization" you said referred to. For more related information, you may search Normalization in the original videopose3D repo.

Sorry for the late response.

zh-plus avatar Jul 27 '21 08:07 zh-plus