Why Toys Fear a Man’s Imagination: How to Make an AI Tsunami Scene From a Toy (Step-by-Step Prompts)
Imagination · CV-002
Why toys fear a man’s imagination
First you see the real footage: a man looking at a Buddha toy. Then you see what he imagines: the Buddha in the middle of Mount Everest, and a giant wave rising behind the mountain. The first part is filmed. The second part is AI. Here is how we made it.
- Format16:9 widescreen
- TypeCustom video
- MethodReal footage + AI imagination
The shot is coming soon
The first part is real footage. The imagination is AI-generated.
That is the shot we made. Below we go through it step by step, so you can see why every prompt is there. Every prompt is at the end, ready to copy.
The recipe
The idea in one line
Real footage + a reference sheet + a first frame + one big action. That is the whole recipe. The steps below show how to put it into words.
Before you start
What you will learn
- How to open a video with real footage and cut to an AI “imagination”
- How to turn photos of a toy into a reference sheet
- How to place that object in a big setting
- How to describe a wave that rises to a set level
- A toy and a camera (a phone is fine)
- A video editor to join the two parts
- An AI image tool and an AI video tool that accept reference images
- A few tries, so you can compare versions
Tool features and settings change, so check your own screen as you follow along.
Step by step
How we made the video
Step 01
Film the real part
Start with real footage, not AI: you, looking at a Buddha toy. Keep it simple and calm. This is the “before” that makes the imagination feel big. Film the toy from the angle you will use in the next steps.
Step 02
Make a reference sheet of your toy
Before the scene, turn the attached photos of your toy into a clean reference sheet on a plain background, so the AI sees one clear version of the statue.
Our thinkingA plain background keeps the shape, the colour and the details of the statue easy to read, so it is easier to keep the same in later images. In the imagination, the toy can stand tall on the mountain.
Step 03
Make the first frame
Now make the picture the video will start from: what the man imagines. Attach the reference sheet and place the Buddha in the setting.
Short and plain: the subject, and where it is. Add light, weather or camera words if you want more control.
Step 04
Turn the image into a video
Set the aspect ratio to 16:9 in your video tool, attach the first-frame image and describe one big action: where the wave comes from, and where the water stops.
Our thinkingA wide 16:9 frame suits this shot, so the mountain, the wave and the statue all stay in view. We name three things: the wave, where it comes from (behind the mountain) and where it stops (the neck). The statue is the anchor of the shot.
Step 05
Join the real part and the imagination
In your video editor, put the real footage first and the AI video right after it. Cut on the eyes or on the toy, and match the angle and the light so the jump feels like a thought. Add a short title text or sound if you like.
Step 06
Generate a few versions and pick the best
AI video is rarely one and done. Generate a few versions of the same prompt, compare them, and keep the best. If the statue changes, attach the reference sheet again. Then join the best AI clip to your real footage.
If it goes wrong
Ideas to try, one change at a time
The statue does not match your reference
The wave crashes instead of rising
The statue is washed away or breaks
The mountain does not look like Everest
Copy and paste
The full prompts
Attach your images where the prompt says so, then paste each prompt in order.
Create Product Reference Sheet for AI Video from Attached Images with plain backgroundBuddha statue in the middle of Everest MountainCinematic video of a tsunami coming from behind Everest mountain and swallowing the Buddha statue up to its neckYour turn
What should we imagine next?
Try it with your own toy, and tell us what you get. Tell us in the comments, and follow us for more AI video recipes.
The imagined part of this video was made with AI. The first part is real footage.