Category: Product Photography, 11 min read
The ghost mannequin effect is a garment photo that looks like it is being worn by someone invisible: filled sleeves, a shaped collar, and no mannequin, hanger or model anywhere in the frame. A mannequin holds the garment's shape during the shoot and is removed afterward, and AI tools now produce the same result from a single photo. The part most guides get wrong is where it goes. Amazon's own apparel imaging guide lists ghost mannequins among the things a main image must not contain and asks for adult apparel on a model, so the shot belongs in your secondary slots on Amazon and can be the hero everywhere else.
The ghost mannequin effect is a product photo that makes a garment look like it is being worn by an invisible person. A mannequin, which is a stand-in body form used to hold clothing in its natural shape during a shoot, gives the garment its shape while the photo is taken. Then the mannequin is removed digitally. What is left is a three-dimensional garment floating in the frame: filled sleeves, a shaped collar, no hanger, no flat surface, no person.
This is written for somebody who has just met the term, not for a seller who already runs the format and wants the operational detail. The full ghost mannequin playbook covers equipment, costs and per-platform workflow for that reader. This one starts at zero and defines every term the first time it appears.
There is one thing this guide does differently from most explanations of the format, and it is the reason it was rewritten. Almost every article about the ghost mannequin effect tells Amazon sellers it satisfies Amazon's clothing main image rule. Amazon's own imaging guide says the opposite, in a list, in writing. That section is below, with the quote.
Take the words separately. A mannequin is a body form, usually plastic or fabric over foam, used to hold clothing in its worn shape during a photo shoot instead of putting the item on a live person. Ghost describes what happens after the shoot: the mannequin is edited out, so the garment appears to hold its shape with nothing visibly inside it.
The same technique goes by several names, all meaning the identical result: invisible mannequin photography, the hollow man effect, and neck joint photography. If a retoucher quotes you for a neck joint, that is this.
That third name points at the hard part. The neck joint is the collar opening left behind once the mannequin is gone. Fill it wrongly and the image reads as a garment with a hole cut in it rather than a garment somebody just stepped out of. Reconstructing that opening, so the inside of the collar and the shoulder seams look like the inside of a real garment, is the whole craft of the format and the reason it used to be expensive.
It is also the tell for whether a ghost mannequin image is any good. Look down the neck opening. If you can see the inside back of the collar and it is lit rather than a black hole, the image is doing its job.
Sellers reach for three alternatives instead: a flat lay, a hanger shot, or a model. Each is the right call sometimes, and each gives up something specific that the ghost mannequin format keeps.
A flat lay is shot from directly above with the garment laid flat. It is quick and it is honest about color and pattern. For anything with real construction it fails at the one job that matters, which is showing what the garment does when a body is inside it: the shoulders collapse, the collar lies flat, and a tailored jacket reads as a rectangle.
A hanger shot at least shows drape, because gravity is doing something. The interior stays empty, the shoulder line takes the shape of the hanger rather than the shape of a person, and the neck and chest stay undefined.
A model is the most informative of the three and the most expensive, and it introduces a variable the other formats do not have: the person. A face, a pose and a haircut all compete with the garment, and a buyer who does not identify with the model can read the wrong thing about the clothing. The ghost mannequin format keeps the fit and structure information a model gives you and removes the person from the decision.
| Format | Shows worn shape | Shows the inside | Adds a person | Typical cost driver |
|---|---|---|---|---|
| Flat lay | No | No | No | A surface and a light |
| Hanger shot | Drape only | No | No | A hanger and a light |
| Ghost mannequin | Yes | Yes, through the collar | No | Two shots and a composite |
| Live model | Yes | No | Yes | A model, a studio and a day |
A comparison of what each format communicates, not a ranking. A flat lay is the right choice for a printed graphic tee and the wrong one for a tailored coat.
The studio method has not changed much in twenty years, and knowing it is what makes the AI version legible.
The cost is the point. Two setups, two shots and a composite per garment means a brand with 50 pieces is buying 100 frames and 50 retouches before anything goes live, and repeating a meaningful share of it every season. That arithmetic is why the format stayed a big-brand thing for so long, and why it stopped being one.
Here is the part almost every guide gets backwards, this site's earlier version of this article included.
Amazon publishes a Seller Imaging Guide for Clothing & Accessories. On its main image requirements page, under the heading of what a main image must not contain, it lists visible mannequins and ghost mannequins side by side. On the facing page, under what a main image must do, it states: all adult apparel should be imaged on-model, and all kids, baby, accessories, multipacks and sets should be off-model. The consequence is printed underneath in capitals: failure to meet these requirements may result in the removal of your ASIN from search.
That looks like it contradicts the guide's own glossary, which lists a ghost mannequin as one of the off-figure image types alongside a flat or laydown shot. It does not. Off-figure images are variants, meaning the alternate slots behind the main image, and Amazon allows up to six of those for seven images in total. The format is welcome on the listing. It is not welcome in the first frame of an adult apparel listing, and that is the distinction the guides collapse.
Two honest caveats before you act on it. The edition Amazon hosts is Spring 2021, and every page carries the line that the guide is for informational purposes only. Your own subcategory's style guide in Seller Central is the authority, and the apparel subcategories are the ones that vary most. Read yours once rather than applying a single rule across a catalog. The general technical spec, which applies whichever slot you are filling, is in our checklist of Amazon image requirements.
Amazon is the strict one. Everywhere else the ghost mannequin shot can be the first thing a buyer sees, and on one marketplace it is close to being asked for by name.
Etsy publishes no rule about how a garment must be presented. Its image guidance is technical: at least 635 pixels on the first photo to avoid ranking lower, at least 2000 recommended, and no transparent PNG, because Etsy renders transparency as black. A ghost mannequin shot flattened onto a real background is a strong Etsy thumbnail precisely because it reads as a product rather than a lifestyle photo at 200 pixels wide.
Shopify has no background or presentation requirement at all, which means the only standard is the one your own store sets. A catalog shot consistently on the ghost mannequin format looks like a brand. A catalog that mixes flat lays, hanger shots and two different models looks like a marketplace.
Mercado Libre is the interesting one, because its packs of the same product page asks for the units on a transparent mannequin, which is this format under a different name. It is one of the very few places a marketplace specifies the technique in its own requirements rather than leaving it to the seller. The rest of its rules, which change by category, are in our guide to Mercado Libre photo requirements.
| Platform | Main or first image | Secondary slots |
|---|---|---|
| Amazon, adult apparel | On-model. Ghost mannequin is in the main image don'ts | Yes, as an off-figure variant |
| Amazon, kids, baby, accessories | Off-model, per the same guide | Check your subcategory style guide |
| Etsy | Allowed. No presentation rule is published | Allowed |
| Shopify | Allowed. No presentation rule exists | Allowed |
| Mercado Libre, packs | A transparent mannequin is what the page asks for | Allowed |
Amazon rows from its Seller Imaging Guide for Clothing & Accessories, Spring 2021 edition, read 16 September 2026; the guide is marked informational and your subcategory style guide in Seller Central governs. Etsy and Shopify rows mean the absence of a rule, which is not the same as an endorsement. Mercado Libre row from its packs of the same product photo requirements page.
Not everything earns it. The format is paying for structure, so it pays off most where structure is what the buyer is trying to judge.
Structured pieces benefit most: blazers, jackets, coats, suits, tailored trousers and any shirt with a real collar and a proper shoulder. These have construction that simply does not read any other way, and a flat lay of a blazer is close to useless.
Knitwear benefits nearly as much, for a different reason. Ribbing, cable and gauge all look like what they are once the garment is filled out and look like texture on a rectangle when it is not. Dresses and skirts benefit at the neckline, the shoulder and the waist, where the fall of the fabric is the thing being sold.
T-shirts, lightweight tops, activewear and swimwear benefit least. Their appeal is how they move and sit on a real body, which a rigid worn shape cannot show, and these are the categories where a model shot earns its cost. Virtual Model covers that case from the same source photo, without a booking.
The barrier was never the idea. It was owning a specialist mannequin with removable sections, shooting twice, and paying somebody to composite the neck joint without it looking cut out.
AI generation removes all three. One flat lay or hanger photo of the garment goes in, and the model works out the worn shape, fills the sleeves and body, and reconstructs the inside of the collar, in one step and with no editing skill involved. The hero image at the top of this article is exactly that: the left frame is a folded sweater on a grey surface, and the right frame is what the same file became.
The practical shape of a listing, then, is two outputs from one garment rather than two shoots. A model shot for the Amazon main slot, a ghost mannequin shot for the secondary slots on Amazon and for the hero on Etsy, Shopify or your own store. AI Product Photography handles the backgrounds and scenes around both.
Each image is one credit, about 12 cents on the $12 Starter plan, and every plan including the free one starts with no card. For a 50 piece collection that is the whole ghost mannequin set for less than the cost of retouching two garments the traditional way. Shotova pricing has the detail.
Shotova is an AI canvas for people selling clothing and products online. The idea is that one photograph of an item should be enough to produce the whole listing, so floating-garment shots, on-model images, angle sets, clean backgrounds, a vertical video and the written title and description all come off a single board.
For a clothing seller that shape matters more than it sounds, because apparel is the category where the platforms disagree most: one marketplace wants a model in the first frame, another wants a transparent mannequin, a third wants whatever you think is best. Producing both looks from the same upload is cheaper than picking one and hoping. The free plan is 10 credits a month with a commercial license, and Shotova Canvas runs a full collection through in one pass.
One photo of the garment, both the model shot and the floating one. 10 free credits every month, no card needed.
It is a product photo technique where a garment is shown in its natural worn shape, floating in the frame, with the mannequin that gave it that shape removed digitally afterward. No mannequin, hanger or person appears in the final image, only the garment. It is also called invisible mannequin photography, the hollow man effect, and neck joint photography, and all three describe the same result.
For adult apparel, generally no. Amazon's Seller Imaging Guide for Clothing & Accessories lists ghost mannequins among the things a main image must not contain, alongside visible mannequins, and states that all adult apparel should be imaged on-model. The same guide treats a ghost mannequin shot as a legitimate off-figure variant, so it belongs in your secondary slots. Kids, baby, accessories, multipacks and sets are told to be off-model instead. Check your own subcategory's style guide in Seller Central, because the apparel subcategories vary.
A flat lay is shot from above with the garment laid flat on a surface, which loses all three-dimensional shape: the shoulders collapse and the collar lies flat. A ghost mannequin photo preserves the worn shape, with filled sleeves, a structured collar and a visible interior through the neck opening, and no visible support. For a structured garment the difference is the whole purchase decision; for a printed t-shirt it is close to irrelevant.
No. AI ghost mannequin tools generate the floating garment result from a single flat lay or hanger photo, reconstructing the worn shape and the inside of the collar computationally. Owning a specialist mannequin with removable sections is what the traditional two-shot studio method requires, and it is the main reason the format used to be out of reach for small brands.
No, and treating it as a default wastes money. It pays off where construction is what the buyer is judging: blazers, jackets, coats, tailored trousers, collared shirts, knitwear, and dresses at the neckline and waist. It pays off least on t-shirts, lightweight tops, activewear and swimwear, where how the piece moves on a real body is the thing being sold and a model shot does more work.
Because it is the part of the illusion that breaks. When the mannequin is removed it leaves an opening at the collar, and if that opening is not filled with a believable interior the image reads as a garment with a hole cut in it. Filling it convincingly, so the inside back of the collar and the shoulder seams look continuous, is the hardest step in the traditional workflow and the fastest way to judge whether any ghost mannequin image is good.
Amazon.com, Inc. (2021). Seller Imaging Guide: Clothing & Accessories, Spring 2021. Retrieved September 16, 2026, from https://m.media-amazon.com/images/G/01/SPIS/Fashion_Apparel_Imaging_Guidelines_Spring_2021.pdf
Etsy, Inc. (2026). Requirements and Best Practices for Images in Your Etsy Shop. Etsy Help Center. Read September 16, 2026, from an Internet Archive snapshot dated July 20, 2026, of https://help.etsy.com/hc/en-us/articles/115015663347
Shopify, Inc. (2026). Product media types. Shopify Help Center. Retrieved September 16, 2026, from https://help.shopify.com/en/manual/products/product-media/product-media-types
Mercado Libre Global Selling. (2026). Photo requirements to sell Packs of the same product. Retrieved September 16, 2026, from https://global-selling.mercadolibre.com/learning-center/news/photo-requirements-to-sell-packs-of-the-same-product