You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Great Work!
Following README, I found you control the view by spherical coordinate system.
So can we sample from certain viewpoint with relative pose metric R,T ,and how to make it?
Thanks!
The text was updated successfully, but these errors were encountered:
Hi,
If you have a custom video and wish to run the model with specified camera rotation and translation parameters, I would recommend running infer.py as documented here: https://github.com/basilevh/gcd?tab=readme-ov-file#custom-data-no-gt
You do not need to know the absolute spherical coordinates (of either the source or destination viewpoint) to do this, since the model is trained with relative camera pose instructions only.
Please let me know if this helps!
Great Work!
Following README, I found you control the view by spherical coordinate system.
So can we sample from certain viewpoint with relative pose metric R,T ,and how to make it?
Thanks!
The text was updated successfully, but these errors were encountered: