How to convert prediction to world coordinates?

una-dinosauria / 3d-pose-baseline

A simple baseline for 3d human pose estimation in tensorflow. Presented at ICCV 17.

MIT License

1.42k stars 356 forks source link

Closed alecda573 closed 4 years ago

alecda573 commented 4 years ago

How do you guys convert the prediction from camera coordinates to world coordinates when using your model on in the wild videos?

una-dinosauria commented 4 years ago

Hi @alecda573,

We used the camera parameters of Subject 11, camera 1, to create our videos in the wild. Please see https://github.com/una-dinosauria/3d-pose-baseline/issues/28#issuecomment-332710381 for more details.

Cheers,