Amazon clothing image requirements: MAIN photos and AI models

For ordinary adult clothing on Amazon US, prepare a MAIN image showing the garment on a standing model against pure white. Keep the whole garment visible. Treat a street scene as a separate creative image, then check the additional-image rules before uploading it.

If a photorealistic person was entirely generated by AI, Amazon also requires a specific metadata tag in the image file. An attractive result and a large export do not establish listing eligibility. Decide the image role, verify the product and prepare its metadata before upload.

This guide follows the current US Amazon product image guide and the clothing-specific rules linked below. Other marketplaces and product types can differ.

Prepare the clothing MAIN image

Amazon identifies the first listing image with the MAIN variant code. Its clothing guidance calls for a frontal, complete view on pure white, with the product occupying about 85% of the frame. Avoid visible hangers or mannequins, folded garments and distracting accessories.

Use the correct colour for each child product as well as a parent MAIN image. A pile of colour options does not replace those individual images. Adult clothing generally needs an on-model view; accessories and certain multipacks have different presentation rules. Check the exact category before preparing the file.

Use the current clothing image guidelines for exceptions, including restrictions on children’s form-fitting clothing. Do not apply an adult-clothing example to those products.

Place the denim images deliberately

The complete jeans reference below identifies the input product. It is useful for checking the generated waistband, pockets and wide leg. A flat product reference alone does not meet the adult-clothing MAIN model requirement.

MAIN

Prepare a separate qualifying front view on a standing model and pure white. Use reliable product evidence for the actual colour, construction and item being sold. Neither street result at the end is a MAIN candidate because both have a location background.

Additional wearing view

The full-length street result shows an outfit direction. Consider that role only after checking garment accuracy, permitted styling and synthetic-person metadata. The coat and boots must not create confusion about what is included.

Additional detail view

The closer result makes the waistband and stitching easier to inspect. It does not establish exact fabric, fit or craftsmanship. Keep a real detail photograph when those facts drive the purchase decision.

Prepare the AI-person disclosure tag

For photorealistic people entirely generated by AI, Amazon requires contains-synthetic-performer as an item in XMP dc:subject, specifically inside rdf:Bag. Its stated trigger excludes images without people and real people merely altered with AI. The tag does not waive any other image requirement.

Follow the current Amazon metadata instructions. Work on a copy of the final downloaded image and preserve its original metadata.

A keyword visible in an editor is only the first check. On our JPEG copy, Preview saved the keyword in rdf:Seq. That differed from Amazon’s documented rdf:Bag structure. The ExifTool command below wrote the required structure without changing the image pixels.

With ExifTool available, replace the quoted filename with your own final file. Read the current subject first. Append the keyword only once, then inspect the saved XMP packet.

exiftool -G1 -XMP-dc:Subject "denim-additional-image.jpg"

exiftool -XMP-dc:Subject+=contains-synthetic-performer "denim-additional-image.jpg"

exiftool -b -XMP "denim-additional-image.jpg"

Check that dc:subject contains rdf:Bag with an rdf:li value of contains-synthetic-performer. Recheck after any conversion or export. A successful metadata write is not an Amazon listing approval.

Handoff: additional wearing view, dark indigo jeans, entirely AI-generated person. Disclosure field: XMP dc:subject / rdf:Bag. Still to review: accurate garment, category rules, permitted styling and uploaded listing.

Check the export and uploaded listing

The general US guide accepts JPEG, TIFF, PNG and non-animated GIF, with 500 to 10,000 pixels on the longest side and at least 72 dpi. JPEG is recommended. Clothing guidance calls for at least 1,000 pixels in either dimension for zoom. Check the downloaded file rather than relying on a quality label.

Check the file’s actual dimensions, colour, sharpness and metadata. Avoid artificially enlarging a small image. After upload, inspect the correct parent and colour variation, then review the listing at a small screen size. Amazon may select, arrange or modify submitted images; upload is not a guarantee of display.

Keep the current technical and all-image requirements beside your final check. Product accuracy, prohibited graphics and content restrictions still apply to additional images.

Use Fotogenic AI for the creative scene

For a completed Media-source image in the Fotogenic AI WordPress plugin, open Summary and choose Open image when the result is Saved. Save the opened file locally, then prepare its role and metadata. This opens the generated attachment; it does not submit it to Amazon.

The two Fotogenic AI results below come from the same jeans reference and serve different creative roles: a wearing view and a closer view. They show a visual direction, rather than Amazon approval or measured fit.

For the scene-production steps and an adaptable location brief, follow the lifestyle clothing workflow. Keep that creative task separate from the MAIN-image and disclosure decisions in this guide.

For the broader channel use case, explore Amazon product photography.

To prepare a plain starting image, follow the white-background guide. A white-looking background alone does not certify a clothing MAIN image.