
Oryginalny dokument
GPT Image 2 · en
Scene:
Create a square luxury retail campaign image staged like a flagship holiday window at blue hour. Use warm amber bulbs, soft snow haze outside the glass, deep spruce green, ribbon red, and brushed brass reflections. The camera should sit at a natural shopper eye level, with enough depth to feel like a real storefront but enough order that the frame reads as a polished hero asset for premium merchandising and gifting communication.
Subject:
Center stacked gift boxes, a hero hamper, evergreen garlands, velvet ribbons, matte ceramic ornaments, and a small amount of bakery or fragrance gifting cues arranged into one coherent Christmas retail composition. The objects should feel premium and tactile rather than toy-like. Build clear foreground, midground, and background layers so the viewer immediately reads wrapping quality, bundle hierarchy, and a believable in-store campaign language.
Important details:
Lock the campaign grammar here: pine texture, ribbon thickness, brass hardware, cream paper stock, amber lighting direction, and a refined balance between abundance and breathing room. Keep the gift architecture elegant and commercially usable, with no cluttered department-store chaos. Surfaces should feel physically real, gift-ready, and suitable for a reusable retail toolkit that could expand into web banners, window displays, and catalog covers.
Use case:
Christmas Retail series study 1/9.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, no character mascots, no cartoon sleigh clichés, no busy crowd, no branded packaging, no readable price cards, no excessive snowfall blocking the products, no cheap glitter explosion.Jak używać
- Skopiuj cały oryginalny prompt wraz z parametrami.
- Dostosuj temat, tekst i kompozycję do swoich potrzeb.
- Użyj zgodnego narzędzia zewnętrznego; sprawdź warunki dostępu i wynik.
Przed rozpoczęciem
- Sprawdź, czy potrzebny jest obraz referencyjny. Zgodność nie potwierdza modelu użytego do przykładu.
- Sprawdź nazwy, liczby i napisy; wyniki zależą od wersji modelu.