Dawood Zahid: Organize an AI Product Video Around Two Reference Images

Use the product-image-plus-person-reference concept to plan assets, reference roles, motion and product fidelity, with a segmented overseas product-video exercise.

“Can I make a video with just one product photo?” is a common seller question. This 10-second Dawood Zahid video makes the inputs concrete: one product image and one character image.

The upper part shows a long-haired woman in white speaking at a table. Below it, reference images of green product bottles and the character appear side by side, followed by a prompt area. The page carries an AI-generated-content label, and the caption explicitly describes the two-image input.

This layout lets viewers compare inputs and output immediately. Engagement on this particular example is modest; its value here is the way it organizes and presents source material. The exercise below uses your own product. Two images are a starting point, and the finished video still needs review and revisions.

What to examine while watching

Compare the product reference with output side by side: outline, color and la…

Try this example

I want to organize a product video using two references.
Image one: the actual product. Image two: an authorized person or original fictional character.
Fixed product facts: [structure, color, markings, dimensions]. Use case: [fill in].
First assess whether the references are sufficient and list hidden or unconfirmed details. Then propose three short shots with one action each, specifying what each reference controls.
Provide Chinese shot notes, English generation descriptions, real-footage additions and a review checklist. If the selected tool lacks multiple-reference support, use a composed and verified starting frame instead. Do not invent interface buttons.

Membership unlocks the full article. Sign in