Oryginalny dokument
GPT Image 2 · en
A realistic-style selfie taken in a high-end restroom with a sophisticated ambiance, against a dark background.{argument name="wall color" default="Dark red"} The textured wall features a large, round mirror with a thin black frame at its center. A person with long {argument name="hair color" default="Dark brown"} Wavy hair{argument name="subject" default="young women"} She stands slightly off-center, her upper body leaning forward and to the side, exuding a relaxed and confident posture. One hand holds a phone, partially obscuring her face, while the other rests on a dark stone or marble countertop. She wears a black, fitted, off-the-shoulder long-sleeved top with a heart-shaped neckline, paired with high-waisted charcoal gray or black jeans, and carries a black shoulder bag under her arm. The phone flash is extremely bright, completely obscuring her face and creating a strong starburst effect in the image, along with a distinct blue-white horizontal lens flare, accompanied by subtle halo effects and a realistic overexposure reminiscent of mobile photography. Behind her is a dark wood-paneled door reflected in a mirror. The room includes two tall, cylindrical wall lamps with warm bulbs, positioned on either side of the mirror's edge; a small gold light switch panel on the left wall; a curved, brushed metal faucet at the bottom center; and two small candlesticks or prayer lamps along the lower right edge. The style is realistic mobile photography, with slight graininess, warm low-light environment, realistic reflection effect, slight sensor noise, and natural shadows, presenting the private nightlife bathroom aesthetics, and using vertical portrait composition.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.
