What is image-to-video in AI video?

Image-to-video animates a still image into a clip. You provide the image plus a short prompt that adds motion and one camera move; the image already supplies the subject, wardrobe, and setting. Kling 3.0 Omni is the image-to-video leader in 2026.

What is video-to-video editing?

Video-to-video rewrites an existing clip from a plain-language instruction while preserving its motion, timing, and camera path. With Gemini Omni you describe only what should change (location, subject, dish, style) and a single input clip can produce many edited outputs.

What is the six-block formula for an AI video prompt?

A reliable structure for a single-shot, text-to-video prompt: lead with the camera (one move), then subject, action, environment, style and lighting, and quality constraints, one short sentence each. Aim for 60 to 100 words.

Which AI video models matter in 2026?

Three: Seedance 2.0 for multi-reference and multi-shot scenes, Kling 3.0 Omni for image-to-video, and Gemini Omni for conversational video-to-video editing.

How do you turn a storyboard into a video?

Generate a multi-panel storyboard as a single image so the character stays consistent, then animate it into one continuous scene with a reference-to-video model such as Seedance 2.0.

#Foundations · Free field guide

The basics of
AI video prompting.

From one image to a finished scene, and back to the simple formula that makes it all work.

A short, practical primer from the upcoming AI Video Playbook 2026. Every clip on this page was made for it: image-to-video on Kling 3.0 Omni, video-to-video on Gemini Omni, the story scene on Seedance 2.0.

01 The payoff 02 Image → Video 03 Video → Video 04 Build a prompt

01Start with the shortcut

A storyboard becomes a scene.

Sixteen panels, one image generated with GPT-Image 2, animated into a single continuous fifteen-second scene on Seedance 2.0, with sound. Below, we break down the building blocks that get you here.

A 4x4 storyboard: a man eats spaghetti, steps out for a call, crows raid his plate, the cat fights them off, he scolds the cat, then realizes and hugs it. — The storyboard: 16 panels, one image, one consistent character.

Play the animated scene — The animated scene

02Low-hanging fruit · one

Image → Video.

The most practical method there is. Your still already does 80% of the work. It shows the subject, the wardrobe, the world. Your prompt only has to add motion and one camera move.

The source still: a man in a yellow beret eating spaghetti in an elegant restaurant, a white cat seated behind him.

The input still

Same frame brought to life using nine different camera motion terms generated with Kling 3.0 Omni.

Same still. One word each. Nine different shots. Tap any clip to play it large, then use ← → to browse.

Specifying the shot type is the single most impactful thing you can add to a prompt, and the cheapest way to learn it is to shoot the move on your phone once.

03Low-hanging fruit · two

Video → Video.

Not a filter, a rewrite. Your clip provides the skeleton: motion, timing, camera path. Your prompt provides the flesh: subject, world, style. With Gemini Omni you just say what should change, and fan one clip out into many.

One input, many outputs

Take one clip and keep talking to the model. Each edit below is a single sentence: the motion and camera stay identical, only what you named gets rewritten.

These six outputs have sound. Tap to play with audio.

Edit: change the location — New location

Edit: spaghetti to pizza — Swap the dish

Edit: change the setting to a train — Change the setting

Edit: to Mars eating rocks — Whole new world

Edit: swap their positions — Swap the character

Edit: bring the cat next to the man — Move a character

Same input video in every panel. The framing, camera path, the character and the world change according to how you describe it.

Now the craft

It all starts as a sentence.

Every clip above began as a few words. The prompt got smaller and the context got larger, but structure still wins. Here is the formula, and a constructor to build one yourself.

04The foundation

The prompt formula.

Structured prompts win. Six short blocks, in order. Build one below: pick or type each term, and it adds up to a ready-to-shoot prompt.

Try it · prompt constructor

Pick from a menu or type your own in any field. The prompt builds itself in the six-block order, and the final prompt is fully editable before you copy it.

This builds a single-shot, text-to-video prompt. Multi-shot scenes, image-to-video, and video-to-video each follow different rules. The full playbook, in the works now, will cover all four. The newsletter gets it first.

01 · Camera

02 · Subject

03 · Action

04 · Environment

05 · Style & light

06 · Constraints Quality guardrails

Your prompt (editable)

Aim for 60 to 100 words, one action, one camera move. Under 30 is under-specified. Changing a field above rebuilds the prompt.

The full AI Video Playbook is in the works. Get it first, plus one tested workflow a month.

Bring your idea to life.

The craft above is just the start. The real shift is context engineering (giving the model the right context to work from), and the part that never gets automated is your direction. The newsletter gets the full playbook first, hands-on training brings it to your team, and the live masterclass covers what actually works.

Train your team Partner on AI video Join the masterclass

Creativity is human. AI is the amplifier.