Images & Video · Podcast episode · source Nov 8, 2025
Product Videos from Lifestyle Images with Custom GPTs and Gemini
Turn existing product and lifestyle photos into short product videos
He showed failures alongside wins, such as a hand reaching in awkwardly and a person who didn't appear in a mirror, and said usable clips take iteration.
Per source From a vendor-affiliated source (see Worth knowing), not independently verified.
Copy-paste prompt
Here's a lifestyle photo of my [product]. Write 5 short video prompts (10-15 seconds, portrait) that keep this exact scene and product, each with one simple motion, lighting direction and camera behaviour. Don't change the product's shape, colour, label or size.
The source didn't publish its exact prompt. We wrote this starter so you can try the same workflow. Replace anything in [brackets].
What they did
On a Helium 10 webinar, an AI trainer for Amazon sellers showed two custom GPTs: one turns a lifestyle image into five motion prompts that stay inside the scene, the other writes a full 15-second shot list (format, lighting, camera notes, four-second beats) from a product photo or description. He ran the prompts in Sora 2, and showed Gemini's Veo making a video straight from a lifestyle image.
How it went
He showed failures alongside wins, such as a hand reaching in awkwardly and a person who didn't appear in a mirror, and said usable clips take iteration. In Gemini the image-to-video step has to start in a new chat.
We only report what the source says. Results are self-reported unless noted.
Worth knowing
Helium 10-hosted and tool-affiliated, with no performance data. Useful as a realistic picture of AI video: expect several runs per usable clip, and check the product looks right in every frame.