One Shoot, Many Markets: Fashion AI

A global fashion retailer ships the same wrap dress to customers in Tokyo, Riyadh, and Lagos. Each of those customers wants to picture themselves in that dress before they click buy. Traditionally that meant either shooting the same product three times with different models, or compromising and showing one model to everyone. Both options leave money on the table.

This is the problem content enrichment solves. Take one core product shoot and multiply it across markets, model types, and contexts without booking another studio. In this post we'll walk through the exact workflow: start with a flat-lay product, generate a base campaign on one AI identity, then localize that campaign for three different regional markets, all in a single afternoon.

The enrichment workflow

The chain has two stages, both running on the On-Model platform:

1. Flat-to-Model turns a single product flat-lay into multiple on-model shots with the same identity, locked to a consistent PDP photographic style.
2. Model Swap takes those base shots and re-runs them with different identities, preserving every garment detail.

Stage one gives you a cohesive base campaign. Stage two enriches it across regional markets. Together they turn one product image into a full localized campaign, without coordinating a single human photoshoot.

Step 1: The base campaign

We start with a single flat-lay outfit: a forest green wrap dress with a printed silk scarf and white sneakers. This is the kind of asset most brands already have in their PIM, often a product team's quick capture of a season's hero look.

For the base campaign we picked Sofia, a Pro-tier identity with a soft, classic look. She'll be our default model, the version of this campaign that goes out before any market adaptation kicks in.

We sent the flat-lay to flat-to-model with three PDP-style instructions, all sharing the same cream studio backdrop, the same soft daylight, the same lens and aperture, varying only the pose. Three cohesive shots, one consistent style, one identity. A real photographer would call this a mini editorial. For a deeper walkthrough of how a single-identity multi-pose campaign comes together with presets, see our Urban Streetwear presets guide, which builds a streetwear mini-campaign with the same one-identity multi-shot pattern.

Step 2: Localize the campaign for three markets

Now we have a complete base campaign on Sofia. The next step is the enrichment itself: re-shooting those same three frames with three different identities, one per regional market. We use model-swap, which preserves the garment, the pose, the backdrop, and the lighting, replacing only the person.

We picked three destination identities from the On-Model catalog, each chosen for clear regional resonance. None of these are real people, all are synthetic.

Each model-swap job took the three Sofia base shots as input and produced three outputs with the new identity. Three jobs, nine total enriched shots, each one a faithful localization of the original campaign frame.

Here's the swap step visualized as a flow: the three Sofia base shots go in, and three Leila localized variants come out. Same poses, same garment, same studio, only the identity changes.

The same exact mapping runs against Yuna and Soleil to complete the matrix.

APAC market — Yuna

MEA market — Leila

Sub-Saharan Africa market — Soleil

Look closely at the scarf print, the wrap belt, the white sneakers with the sage suede heel. Every garment detail is identical across all twelve frames (three Sofia originals plus nine localized variations). The only thing that changes is who is wearing the outfit.

Why enrichment matters for global brands

The traditional path to localized fashion imagery involves either re-shooting (expensive, slow, hard to coordinate across regions) or showing the same model everywhere (costs you conversions in markets where shoppers want to see themselves represented). Enrichment splits the difference and gives you both: a cohesive global campaign and a regional adaptation, from a single source asset.

A few patterns we see customers running on top of this workflow:

1. Default plus regional variants. Ship the base Sofia campaign to your default storefront, and serve the localized identity to specific country domains based on geo. PIM systems handle the routing.
2. A/B testing identities by market. Run two or three identity variants per market and let conversion data tell you which one drives the strongest add-to-cart rate. This is impossible with traditional photography because you can't shoot fifty model variants per product.
3. Diverse representation without quotas. Every product gets shot on multiple body types, ethnicities, and ages, automatically. Inclusion stops being a campaign-level decision and becomes a baseline.

The full enrichment math

For one input flat-lay we produced:

• 3 base campaign shots (Sofia, three poses)
• 9 localized variants (3 base shots × 3 destination identities)
• 12 total on-model assets, all consistent, all from one source product

Scale that across a 500-SKU catalog and you have 6,000 on-model images, locked to your brand's photographic style, ready to drop into every regional storefront. The same exercise with traditional photography would take months and a six-figure budget.

What's next

• Brand Identity guide — managing consistent AI model identities across your catalog
• Flat-to-Model presets guide — the full preset system that powered the base campaign in this post
• AI Model Catalog — browse and select from the diverse pool of identities used for localization
• Model Swap guide — the swap workflow that enriches a base shoot into regional variants
• Five input types, one platform — every input format that feeds into the enrichment chain

Ready to enrich your catalog for every market you serve? Sign up, upload one product, and run the full flat-to-model into model-swap chain to see what your global campaign could look like.