Can I use a photo of the garment on a hanger instead?

You can, but a hanger pulls the shoulders into points and hides the side seams, so the model sees a distorted shape. Add a flat lay if you can. Never use a hanger photo as an Amazon main image, because Amazon bans hangers there.

Do I need a new character sheet for every garment?

Not always. A sheet can show the model in a full outfit, and each store-photo prompt swaps in the product. On our trousers job, the prompt kept the sheet’s whole outfit and replaced only the trousers. Remake the sheet when the model’s look has to change, such as a new haircut for a new season.

Which image model should I use?

Test two on the same garment. Our kept trouser store photos came from GPT Image 2, and a kept knit close-up on another job came from Nano Banana 2, but each job changed the model and the shot at the same time, so that is not a ranking. See which AI image model is best for product photos.

How to turn a flat lay into on-model photos with AI

Checked September 24, 20268 min read

Short answer

Photograph the garment flat, front and back, pressed and evenly lit, plus close-ups of the fabric and details. Write down every construction detail and how it should fit. Approve one model on a character sheet, then make each store photo from the sheet and the flat lay, with a prompt that lists the garment’s details and says nothing may be added or removed. Check every photo against the real garment before it goes live.

What you need before you start

Input	Why the model needs it	How to get it right
Flat lay, front	This is the garment the model redraws on a body	Steamed, no loose threads, symmetrical, on a plain surface, lit evenly, shot straight down
Flat lay, back	Without it, the model invents the back	Same surface, light and camera height as the front
Fabric close-up	Weave and texture disappear at full-garment scale	Fill the frame with fabric and light it from a low side angle
Detail close-ups	Buttons, zips, labels, prints and hardware drift first	One sharp photo per detail
Garment facts	A model cannot measure a photo	Fiber, weight, fit (slim, straight, relaxed), rise, where each edge lands on the body, and every construction detail
The size to show	Shoppers compare the model’s size with their own	Pick the size the model wears and print it on the product page

Amazon’s 2021 clothing imaging guide describes a flat lay that also works well as a reference: steamed, free of loose threads, pinned or stuffed to hold its shape, with the seams keeping left and right symmetrical. For a phone setup, see how to photograph your product for AI.

Check what your tool accepts and returns. FASHN’s try-on API takes a flat lay or ghost mannequin photo for tops, bottoms and one-pieces and works at 864 x 1296 pixels. Photoroom’s Virtual Model returns 1024 x 1024 pixels by default, with 2K and 4K on its Enterprise plans. Match the output size to where the photo will be used before you commit to a tool.

Approve the model first

Before any store photo, make a character sheet: one image of the same person from the front, three-quarter, side and back, with face close-ups, on a plain background. Approve it, then send it as the first reference in every shot. If the face is wrong on the sheet, it will be wrong in every photo made from it, so fix it there.

Budget time for it. On our surf apparel job the character sheet ran 20 times before one was kept. The kept one came from 5 well-chosen references and a short prompt; the longest prompt and the attempts with the most references were not kept. The method, including the line that stops labels being printed on the sheet, is in how to keep the same model in every AI photo.

Step by step

Shoot the flat lays and details listed above, and keep the files at full size.
Write the garment spec: every construction detail, including what the garment does not have, the fit, the fabric and its finish, and the colors by name.
Make the character sheet and approve it.
Write one store-photo prompt in this order: who the person is (the sheet), what they wear (the flat lay), the garment detail list, then backdrop, light and camera.
Keep the backdrop, light and camera sentences fixed for the whole set. For each shot, change only the framing and pose, and attach only the references that shot needs: the sheet and the front flat lay for a front view, the sheet and the back flat lay for a back view.
Render two to four versions of each shot and pick one. If one shot is wrong, remake that shot with a note instead of re-running the set.
Check each keeper with the list below, then export at the size your store needs.

The prompt that kept our trousers exact

This prompt made store photos an art director kept on our men’s trousers job, rendered with GPT Image 2. INPUT IMAGE 1 is the approved character sheet and INPUT IMAGE 2 is the flat lay of the trousers. The job took 40 renders for 11 keepers.

Store photo of trousers on a model (GPT Image 2, about 330 words, lightly edited: brand names removed)

This is a photograph of the exact same man as INPUT IMAGE 1, same face, same hair, same skin tone, same build, wearing exactly the same wardrobe head to toe as the character sheet with only the trousers replaced by the product trouser shown in INPUT IMAGE 2. The frame begins above the head and ends below the shoes, the whole figure inside the frame with clean margin, head and shoes fully visible. Full body front view, standing composed and upright with a subtle weight shift resting smoothly on the right leg, arms relaxed at the sides, gaze soft toward camera. The trouser must match INPUT IMAGE 2 exactly, navy smoke houndstooth pattern in a textured woolen weave with a matte finish and substantial drape, flat front with no pleats, angled slash front pockets, standard waistband with belt loops and an extended tab closure, a single dark button at the waist closure, no rivets, minimal tonal visible stitching, sharp center front crease, straight to mildly tapered leg with a plain finished hem, nothing added and nothing removed. If a fabric close-up reference is provided its weave and texture must be matched exactly. Studio setting with the backdrop exactly as follows, Light Gray, a seamless mid-gray studio backdrop with a soft cool tone that grades slightly darker toward the edges and blends into a matching gray floor with no visible seam, lit evenly with diffused soft light so the model casts only a faint soft shadow, no props. Lighting is a large soft key from the upper left at forty five degrees with even wraparound fill, a soft natural contact shadow under the shoes, no hard shadows, no colored light. Camera is a full frame mirrorless body with an 85mm lens at f8, camera at chest height, level, no tilt. Colors stay exactly true to the flat-lay, never shifted, never saturated. Crisp commercial e-commerce finish. Portrait orientation. The garment is never altered from the reference. No text, no graphics, no logos other than what exists on the garment, no watermarks. Natural believable hands and body proportions. Photorealistic only, never illustrated, never stylized.

Three parts hold the garment in place. The trouser sentence lists each construction detail, including what the trousers do not have ("no rivets"), and ends with "nothing added and nothing removed." The color line tells the model to keep the flat lay’s colors instead of grading them to suit the scene. The first sentence keeps the sheet’s whole outfit and swaps in only the trousers.

Across the set, the backdrop and lighting sentences were reused word for word. Only the framing and pose sentences changed from photo to photo, and the camera line changed to a 100mm macro at f/5.6 for the fabric close-up. For prompt structure in general, see how to write an AI image prompt for a product.

Check every photo against the garment

Same number of pockets, in the same places, with the same openings (slash, patch, welt)
Same closure: button count and color, zip color, snaps and tabs
No hardware added: rivets, eyelets, buckles, extra zip pulls
Waistband, belt loops, pleats and creases match the flat lay
Hem finish matches (plain, cuffed, raw), and the length on the body fits the size shown
Print or pattern at the same scale, and stripes meet at the seams the way they do on the garment
Logos and labels spelled right, in the right place, at the right size
Color matches the garment under neutral light, checked on a calibrated screen
Fabric reads the same: weave, sheen, and the weight shown by the folds
The fit looks like the size you list, neither tighter nor looser
Hands and body proportions look natural, with no fingers merging into fabric
Metadata tag added if the person is fully AI-made and the platform asks for it

The full pre-publish list is in the AI image QA checklist.

What to tell shoppers

Print the model’s height and the size worn under the photos, because Baymard’s testing shows shoppers use both to pick a size. If the person is AI-made, follow each platform’s rule. Amazon (US) asks for the keyword contains-synthetic-performer in the image metadata for fully AI-made, photorealistic people, and adds a disclosure where it applies. Google Merchant Center requires AI metadata, such as IPTC TrainedAlgorithmicMedia, to stay in the file. For ads, New York’s synthetic performer law requires a conspicuous disclosure when the advertiser knows one is used.

Questions people also ask

Can I use a photo of the garment on a hanger instead?: You can, but a hanger pulls the shoulders into points and hides the side seams, so the model sees a distorted shape. Add a flat lay if you can. Never use a hanger photo as an Amazon main image, because Amazon bans hangers there.
Do I need a new character sheet for every garment?: Not always. A sheet can show the model in a full outfit, and each store-photo prompt swaps in the product. On our trousers job, the prompt kept the sheet’s whole outfit and replaced only the trousers. Remake the sheet when the model’s look has to change, such as a new haircut for a new season.
Which image model should I use?: Test two on the same garment. Our kept trouser store photos came from GPT Image 2, and a kept knit close-up on another job came from Nano Banana 2, but each job changed the model and the shot at the same time, so that is not a ranking. See which AI image model is best for product photos.

Where Overs fits

Overs runs these steps as a plan. For worn products it makes a character sheet you approve first, then shows the shot plan and an estimated cost before anything is charged. Each photo gets only the reference pictures it asked for; if a needed one, such as the back of the garment, does not exist, Overs skips it instead of guessing with the wrong picture.

Free for 40 photos a month. The AI that makes the photos is billed separately, on your own key, with no markup from Overs.

Start free How it works

Sources

Checked on September 24, 2026. Prices, specs and rules change; follow the links for the current versions.