Lesson plan
What you will build
A finished 30-second cinematic bullet-time ad built from one AI generation, starring a consistent character built from your own photo, with your own branded hero objects and a logo engraving that survives close inspection. You will also have a reusable character-sheet prompt, one complete video prompt you can point at any product, and a frame-level QA method that catches the five failure modes before you publish.
Tools used
- Pinterest (reference sourcing)
- ChatGPT or any image model (character sheet)
- Seedance 2.5 (video)
- CapCut or After Effects (post)
- ffmpeg (QA)
Requirements
- A clear photo of your own face, ideally front-on with even lighting
- Access to an image model that accepts a photo reference (ChatGPT, Nano Banana, Midjourney)
- Access to a video model that takes multiple reference images and generates 20-30s clips
- Product shots of two or three objects you want to feature
- Basic familiarity with a video editor for the final branding pass
Lesson contents and progress
Your progress
0 of 11 steps
Step-by-step class
Follow the lessons in order.
Each lesson includes the exact action to take, the reason it matters, and a checkpoint before you move on.
Lesson 01 of 11
Study the reference before you write a single word
Everyone starts by writing a prompt. Start by counting instead. Download the viral video you're referencing and step through it. You are looking for three things: 1. THE BEAT COUNT. How many distinct moments are there, and how long is each? Most of these ads run 18-25 seconds and contain eight beats: walk, hero reveal, impact, fall, three frozen object macros, and a payoff. The map below is the version you are going to build. 2. WHERE THE CUTS ARE. This is the part nobody checks. Take the difference between consecutive frames. Where the difference spikes, there's a cut. Run this on almost any viral 'one-take' bullet-time ad and you'll find hard transitions every two to four seconds, because it was generated as eight separate clips and edited together. 3. WHAT'S ON THE OBJECTS. Pause on each frozen macro. On the original that started this trend, the creator's handle is laser-etched onto the sunglasses temple arm. That is 2.4 seconds of full-screen branding that reads as product design rather than a watermark. It is the single smartest thing in the video. Why this matters: once you know the reference is stitched, you know you can beat it. A model that generates a full 30 seconds in one pass gives you something the original creator could not have, a genuine unbroken take, which is far more cinematic than eight clips buttoned together.
