A meaningful share of fashion ecommerce brands — boutiques, dropshippers, multi-brand retailers, marketplace sellers — do not actually photograph their own garments. They sell garments their suppliers photograph, and the supplier flat-lays (or supplier model shots) are what they get. The catalog problem is converting these unglamorous supplier inputs into catalog-grade on-model imagery without a studio. AI catalog production is the cleanest path. This is the practical 2026 how-to.
Why supplier flat-lays are the input most brands actually have
The catalog imagery that comes from suppliers (Alibaba wholesalers, domestic distributors, drop-ship partners, white-label manufacturers) is consistently below catalog- grade. Some of it is hanger photography on poor lighting. Some is flat-lay on cluttered backgrounds. Some is on-model but with the supplier's house model whose look does not match the brand's audience. The brand's job is to take this input and turn it into imagery that converts on the brand's own storefront.
Pre-AI, the path to catalog-grade imagery from supplier inputs involved either re-shooting the garment (cost prohibitive for high-SKU low-margin shops) or hiring a photographer for the imagery (same cost problem). Most brands lived with the supplier inputs and the conversion rate they produced. AI catalog production gives a third option: the supplier input becomes the brief, and the AI produces the catalog-grade output.
Step 1: categorise what kind of supplier input you have
The right AI workflow depends on the supplier input format. Common categories: clean flat-lay on a neutral background (best input, easiest to convert); cluttered flat-lay with the garment on a coloured floor or near other objects (needs prep before AI); supplier on-model photograph (different workflow); hanger or mannequin photograph (needs ghost mannequin treatment); product on a black or coloured background (needs background replacement).
Each category routes to a different starting template in the workflow. Brands receiving high volume from multiple suppliers benefit from documenting which suppliers ship which input format and standardising the downstream conversion route accordingly. The discipline upstream pays back across every SKU.
Step 2: clean the input where needed
Cluttered or coloured-background inputs benefit from a background-removal pass before the AI catalog rendering. Run the input through a generic background remover (Photoroom, remove.bg) or through Apiway's ghost mannequin to isolate the garment on white. Use the cleaned output as the input to the next step. Skipping this step causes the catalog rendering to inherit artifacts from the supplier's background.
For supplier on-model photographs where the model does not match the brand's audience, the workflow is different: extract the garment from the supplier model through a virtual try-on operation in reverse, or re-render the garment on the brand's preferred model identity. Apiway's reference-photoshoots and White Studio templates handle this case.
Step 3: render on the brand's locked model identity
The conversion-driving step is rendering the cleaned garment on the brand's locked model identity. The brand's catalog should feel like the brand, not like a portfolio of supplier outputs from different sources. Apiway's White Studio with stable model identity persistence is what makes this operationally feasible across hundreds of SKUs from dozens of suppliers.
Lock the model identity once and use it for the entire catalog. Two or three model identities representing the audience demographic spread is the right answer for most brands. The merchandising team defines which model gets which garment based on category fit and audience signal.
Step 4: add the lifestyle layer through creator marketplace
Catalog thumbnails and PDP main images on the brand's model identity are the foundation. The lifestyle layer through Apiway's creator marketplace is the differentiator. Brands sourcing from suppliers rarely have lifestyle imagery; the supplier ships the product shot and nothing else. Ship the brand's SKUs against creator marketplace photo sets and the catalog acquires a lifestyle layer the supplier never provided.
This is often the highest-impact change for boutique and multi-brand retailers. The catalog goes from supplier-generic to brand-distinct. Conversion lift on most retailers from this single change is meaningful.
Step 5: batch the workflow
Brands receiving supplier inputs in volume benefit from batching the conversion. Standardise the input file format (resolution, aspect ratio, naming convention). Run the cleaning step in batch through the chosen background remover. Run the catalog rendering in batch through Apiway's template. Run the QC pass in batch. The full workflow from supplier input to ship-ready catalog imagery should run as a pipeline, not as ad-hoc per-SKU work.
Apiway supports batch generation natively across the catalog templates. For brands moving 100+ SKUs per cycle, the batch workflow is the operational unlock. For smaller volumes, the manual flow per SKU through the same templates produces the same per-image output quality.
Step 6: QC against the original supplier input
Verify the AI catalog output preserves the actual garment from the supplier input. The most common failure mode at this step is the AI rendering subtly modifying the garment — changing a detail, drifting the colour, simplifying the print pattern. The catalog must show the garment the brand actually sells, not a cleaner version of it.
QC discipline: place the supplier input next to the AI output for each SKU and visually verify garment fidelity. Use the original supplier flat-lay as the ground truth. The QC time per SKU is small but non-zero; brands should budget for it in the workflow.
When the supplier input is too poor for conversion
Some supplier inputs are too low-resolution, too poorly-lit, or too cluttered for AI catalog conversion to produce usable output. The ceiling on the AI output is bounded by the input quality. For these SKUs, the right answer is to either request better input from the supplier (often available on request) or to skip the AI conversion and either drop the SKU from catalog or invest in a single replacement photograph. The AI is a multiplier on input quality, not a substitute for it.
Getting started with supplier-flat-lay catalog conversion
Sign up for a free Apiway account. Pick five recent supplier inputs across different categories. Run them through the cleaning and catalog rendering pipeline on White Studio and Ghost Mannequin. Compare the AI output to the supplier original. The delta between the two is the conversion gap supplier inputs typically leave on the table; the AI workflow closes that gap.
Related reading
See our dropshippers stock-photo plateau essay, our boutique owner AI photos guide, our Shopify clothing photos guide, and the full Apiway blog.