Original document
GPT Image 2 · en
Ultra-realistic portrait of a beautiful Korean woman standing inside an empty Korean classroom during late afternoon golden hour, captured in a cinematic film photography style. She faces slightly to the side with her gaze turning toward the camera, expression soft and introspective — lips gently parted, eyes calm and luminous. Both hands loosely holding the strap of a small canvas tote bag hanging off one shoulder, effortlessly cool posture. Wearing an oversized vintage cream knit sweater slightly off one shoulder, paired with high-waisted dark navy pleated mini skirt and white low-top sneakers. Soft layered gold necklaces, minimal jewelry, understated Korean street fashion aesthetic. Long dark brown hair with subtle highlights, half-tucked behind one ear, loose natural waves. Dewy glass skin, flushed cheeks, barely-there tinted lip, soft eye makeup — natural Korean beauty. Warm amber golden light flooding in through tall classroom windows at a low angle, casting long dramatic shadows across wooden desks and the floor. Dust particles softly visible in the light beams. Authentic Korean classroom setting — rows of wooden desks, vintage chalk-dusted green board with Korean writing, South Korean flag, old wall clock, stacked textbooks, worn polished flooring. 35mm analog film grain texture, Kodak Portra 400 color palette — warm honey tones, muted greens, soft creams. Shallow depth of field, background bokeh on classroom furniture. Nostalgic coming-of-age Korean drama visual tone, quiet emotional storytelling atmosphere, editorial lifestyle photography, cinematic still frame, hyper-realistic skin and fabric texture, golden hour haze, peaceful solitude mood.How to use
- Copy the complete original prompt, including any parameters.
- Replace the subject, text and composition with your own requirements.
- Use a compatible external tool; check its current access conditions and review the result.
Before you start
- Check whether a reference image is required. A compatible model label does not mean this example was generated with that model.
- Keep names, figures and visible text accurate; results can vary between model versions.
