Three short clips generated from three still images.
This is proof that fal.ai can turn your existing stills into moving footage. Each clip below was generated from the still directly above it, using a video model called Veo 3.1 running on fal.ai. Nothing was filmed, and no 3D scene was built.
Give each clip a verdict and write anything you want in the notes box. Your answers save as you go, so you can close the page and come back.
Originals, and what fal.ai made from each one
The top row is the three source stills exactly as supplied, at their own shapes and sizes. Directly under each one is the clip generated from it.
Camera travels straight in toward the front door, closing about half the distance.
How the still was prepared: Used exactly as supplied. It already matched the 9:16 delivery frame, so nothing was cropped or extended.
Click the lit button again to clear it.
Camera runs up the gravel path and carries on through the open frame toward the water.
How the still was prepared: Enlarged to 2048 square, then a 9:16 window was taken from the centre. No part of the picture was repainted.
Click the lit button again to clear it.
Camera moves in toward the pair of glazed doors. The doors stay shut.
How the still was prepared: Trimmed at the sides to hold the window and the doors, then the ceiling and floor were extended to reach 9:16.
Click the lit button again to clear it.