Image-to-Video

A way to make video by giving AI one picture as the starting point, then directing the movement. It's more controlled than text-to-video because the character's look and the background are already set.

Example: A portrait + "camera slowly moves closer, they blink" = a short, consistent video.