Image-to-Video

Get auto detected elevation for sv3d_u

#6
by sam598 - opened

According to Stability staff "sv3d_u automatically detects the elevation in the conditioning image". https://huggingface.co/stabilityai/sv3d/discussions/4#65f99514f5afddb41fba05c8

Unfortunately I cannot reopen the issue.

Is there a way to get the detected elevation? Otherwise creating an accurate model of the camera poses and eventually a mesh does not seem possible, and sv3d_p would need to be used.

Stability AI org

Currently there is no way to get the detected elevation from sv3d_u. Elevation is detected implicitly by sv3d_u

vikky2904 changed discussion status to closed

Sign up or log in to comment