Use Inpainting for Consistency on Single Prompt Animations

I believe that Deforum videos of a single prompt would be more consistent if there was a mode that instead of generating a new frame, inpainted the missing details. For example for a 2D camera that is zooming in what it would do is keep all of the image the exact same (maybe with noise scheduling settings) except for the center which it would inpaint at a strength proportional to the zooming speed, so that the zoomed area would be in proper resolution. For a 3D rotating camera as another example, it would find the areas that MiDaS shows weren't in view last frame.

Start by tracing how single-prompt animations generate each frame and how the 2D and 3D camera paths expose new areas. Inspect the existing MiDaS-related depth handling for the rotating-camera case. Done means a new mode can preserve existing image regions while inpainting newly exposed details with strength tied to zoom or camera movement.

Written by the indexing model from the issue text.