Skip to main content
Ozu is built on a single conviction: AI should accelerate filmmakers, not replace them. The tool exists to get you past the blank page faster, and then to hand you every lever you need to shape what happens next.

​
The thesis

A first-pass storyboard shouldn’t take a week. A working animatic shouldn’t cost a month of labor. The creative work that actually matters (deciding what the scene is about, who’s in it, how it’s framed, how it moves) is the part worth a filmmaker’s time. Everything else is overhead. Drawing boards. Casting extras for a test. Re-shooting a reference because the lens was wrong. The boring parts of filmmaking are boring because they repeat. Those are the parts AI is built to absorb. What AI is not built for is the creative decision itself. That stays with you.

​
The pipeline as a creative tool

Ozu’s pipeline paradigm exists to honor that split. A few principles follow from it:

​
Every creative decision is reversible

Nothing you generate is final until you say so. A frame you don’t like gets reshot. An actor whose look is off gets rebuilt, and every frame that used that actor rebuilds with it. A scene that lands wrong gets re-edited in the Timeline without losing the storyboard upstream. Reversibility is not a feature. It’s a prerequisite for creative work, and the pipeline is designed to keep it cheap.

​
Every asset is reusable

An Actor you generate for scene 1 is the same Actor in scene 12. A Set you build once shows up wherever the script takes you there. Wardrobe variants, lighting moods, camera views: all of it reuses rather than regenerates. This is how the pipeline stays affordable and how your creative choices compound. The work you do early in a project pays off later.

​
Every stage produces something you can read

When the Script Node analyzes your script, you can see every scene, shot, and beat it extracted, and edit them. When the Storyboard Node generates frames, you can open each one and check the inputs it used. When the Timeline assembles an animatic, the sequence is explicit. No black boxes. If something came out wrong, you can always find the step where it went wrong and fix it there.
“One button, full animatic. Then make it yours.” The pipeline is designed so you can drop in a script, press one button, and get a first pass. But the real value is what you do next: reframe, reblock, reshot, re-cut.

​
Respect for the craft

Filmmaking has a vocabulary. Scene, Shot, Beat. Hero Look. Wardrobe. DRB. Framing. Coverage. Ozu uses these words intentionally, because a tool that speaks the craft’s language is a tool a filmmaker can trust. That’s not about nostalgia. It’s about precision. When a director says “close on the hands” and a DP hears “close on the hands,” they’re agreeing on something specific. Ozu aims for the same kind of precision with AI, and the shared vocabulary is how we get there.

​
What we’re not trying to do

A few things Ozu is deliberately not chasing:
  • Photoreal instant feature films. The tools we use are evolving fast, and some day a one-prompt feature may be possible. Today it isn’t, and pretending otherwise would waste your time.
  • Replacing editors, cinematographers, or animators. The pipeline accelerates the planning and visualization phase. Final craft still belongs to humans, and to the editing tools you already use.
  • Locking you in. Every export format (PDF, ZIP, MP4) is a standard file you can open anywhere. Ozu’s job is to hand off clean.

The rooms

A plain-language walkthrough of the pipeline, from script to animatic.

The Studio

Cloud Creator, the studio behind Ozu.

Pipeline and Nodes

The mental model for how work flows through Ozu.

Getting Started

The fastest way to see the philosophy in practice: your first project.

​
FAQ

Is Ozu trying to replace filmmakers? No. Ozu is trying to replace the tedious parts of pre-production, so filmmakers can spend more time on the creative parts. Why not just use a chat interface and one prompt? Because film is structural. Scene, shot, beat, asset, timing: these aren’t optional in a finished piece of work. A pipeline that respects the structure produces work you can actually use; a single-prompt tool produces work you have to throw away. Will the craft arguments hold up as AI gets better? We think so. Better models make the pipeline faster and cheaper, but the creative decisions a filmmaker makes don’t go away. If anything, faster generation means more time spent on choices, which is the whole point. Is this tool for professionals or hobbyists? Both. A pro uses Ozu to compress weeks of pre-production into hours. A hobbyist uses it to see their scenes for the first time. The pipeline works the same either way.