Depth Video Generator | Promptsref ✓

AI Depth Video Generator is an online tool that converts a short source video into a false-color, frame-by-frame depth-map video. You upload an MP4, MOV, or WebM clip up to 30 seconds and 100 MB, and the model estimates relative near-to-far distance across every frame, rendering the result as an H.264 video without audio.

What it keeps is the useful part of a clip: timing, body position, camera movement, and the spatial layering of the scene. What it drops is the original face, clothing, and setting as the main reference, which frees you to point the motion at a new character, outfit, art direction, or environment.

The primary use is as a motion and spatial reference for reference-driven AI video generation, where you reuse a single performance to build many visual versions instead of filming the movement again.

It also works as a depth guide for VFX and 2.5D experiments, such as fog, blur, particles, grading, parallax, and depth-driven motion. Pricing is usage-based at 5 credits per second, rounded up to the next whole second, with the duration and cost shown before generation and credits refunded automatically if a job fails.

It is not a 3D reconstruction: no point cloud, mesh, or camera calibration, just relative depth per frame.