Glossary
What is Lock-and-Outpaint?
Lock-and-outpaint is an AI product photography technique that keeps the product image pixel-locked in its exact original position, crop, and angle while the AI generates only the environment around it. The instruction to the model is explicit: keep the product completely unchanged, do not regenerate or reprocess it, and outpaint the surroundings only, blending in new lighting to match. Because the product itself is never redrawn, the technique removes the drift that occurs whenever an AI reinterprets an object, making it the highest-accuracy route available for product-on-surface photography. From one frozen studio shot, a brand can generate an entire series, the same untouched product placed on wood, silk, concrete, or leather, changing only the surface and the light.
Understanding Lock-and-Outpaint.
The logic is simple: every time an AI model redraws a product, something can drift, a proportion shifts, a logo blurs, a seam moves. Lock-and-outpaint removes that risk entirely by never asking the model to touch the product at all. The prompt instructs the model to keep the product in its exact original position, crop, zoom, and angle, and to treat the surroundings as the only thing open for generation, blending new lighting and shadow onto the frozen product so it reads as part of the scene.
This makes it the best choice for product-on-surface photography and any job where fidelity is non-negotiable and the camera angle can stay fixed. From a single locked studio shot, a brand can generate a whole catalog of variations, the same untouched product resting on wood, silk, concrete, or leather, with only the surface and the light direction changing between them.
The one condition the technique needs is an angle-matched source: it can only lock a product into the pose the source photo already shows, which is why capturing a small bank of product angles up front is the upstream step that keeps this technique available whenever it's needed.
How It Relates to AI Photography.
On the Dezygn platform, lock-and-outpaint is the default route whenever a product's silhouette or fine detail cannot be safely described in words, complex mechanisms, precise hardware, exact stitching, because it guarantees zero drift on the one thing that matters most: the product itself. Full technique breakdown at /resources/lock-and-outpaint.
Related Terms.
Material Prior
A material prior is the AI image model's built-in default belief about what a named material looks like, learned from the training photos it has seen.
Route Map
A route map is a diagnosis-to-technique framework for AI product photography that starts by naming exactly what is wrong with a generated image, using eight fixed accuracy axes, silhouette, proportions, element count, text, graphics, material, color, and construction details, and then selects the one technique built to fix that specific defect.
Count Error
A count error is a product-accuracy defect where an AI-generated image shows the wrong number of a product's repeating parts, buttons, straps, legs, or stitches, even when the rest of the product looks correct.
Start using AI for your product photography.
Turn product photos into conversion-ready visuals with Dezygn's AI Creative Suite.
Start Free