search

vt-vl-lab / 3d-photo-inpainting

[CVPR 2020] 3D Photography using Context-aware Layered Depth Inpainting
https://shihmengli.github.io/3D-Photo-Inpainting/
Other
6.91k stars 1.11k forks source link

dolly-zoom-in arguments #76

Closed sulram closed 4 years ago

sulram commented 4 years ago

Hello there!

I'm curious because dolly-zoom-in have the same arguments of zoom-in. Why the video output is different?

Is there another argument involved?

LemonATsu commented 4 years ago

In dolly-zoom-in, we will recalculate the fov to make the point/object of interest stay at the same size (see: https://github.com/vt-vl-lab/3d-photo-inpainting/blob/e6872a8f7452c71b0f348d3dfcd7c13e1b8b69c1/mesh.py#L2242).

sulram commented 4 years ago

Great, thank you for the great work! @LemonATsu