Glossary

The AI fashion glossary

A plain-language reference for the words used in AI fashion photography and virtual try-on. Each term is defined neutrally, so you can tell an on-model shot from a flat-lay and a garment swap from a ghost mannequin.

AI fashion photoshoot: An AI fashion photoshoot generates on-model clothing images from photos instead of booking a studio, model, and photographer. You upload a garment photo and a model photo, and a model renders the garment worn on the person, in a chosen pose, scene, and light.
Virtual try-on: Virtual try-on places a garment onto a person in an image or a live view so a shopper can see how it looks worn. In e-commerce it usually means putting a catalog garment on a model or on the customer, without a physical fitting.
On-model photography: On-model photography shows a garment worn by a person rather than laid flat or on a mannequin. It sells better because a shopper can read fit, drape, and scale, which a flat image hides.
Flat-lay: A flat-lay is a photo of a garment laid flat and shot from above. It is quick and cheap to make but hides how a piece fits and moves, which is why boutiques convert an flat-lay into an on-model image.
Garment swap: A garment swap replaces the clothing a model is wearing in an image with a different garment, while keeping the same person, pose, and scene. It lets one model photo carry a whole catalog.
Ghost mannequin: A ghost mannequin, or hollow-man, edit removes the mannequin from a garment photo so the piece appears to hold its own shape. It is a catalog convention, distinct from showing the garment on a real person.
Lookbook: A lookbook is a curated set of images that presents a collection as a coherent story, usually one model across several looks. It is used to launch a drop and to give a boutique's feed a consistent identity.
Feed, story, and reel formats: These are the three main image and video shapes on Instagram: the feed post (square 1:1 or portrait 4:5), the full-screen story and reel (vertical 9:16). Posting the right shape avoids awkward crops and keeps important detail out from under the interface.
Safe zone: A safe zone is the area of a story or reel that stays clear of the app's buttons, caption, and profile chrome. Keeping text and faces inside it means nothing important is covered when the post goes live.
Model identity consistency: Model identity consistency means the same recognizable person appears across every image in a set, with stable face, skin tone, and proportions. Without it, a collection looks like several different shoots rather than one.
Composited text: Composited text is a caption, price, or badge added as a separate layer on top of a finished image, rather than drawn into the image by the generator. Because it is a layer, it stays crisp and can be edited later without regenerating the picture.
Upscale: Upscaling increases an image's resolution and sharpens its detail so it is large and clean enough to post or print. A ready-to-post image is upscaled and finished, not a soft low-resolution preview.