GPT Image 2 on ChatGPT
Try with your own image and show the results.
Prompt:
Vertical three-panel cinematic collage, stacked top to bottom, 9:16. Same woman in all three frames, matching the reference image. Consistent facial features, body proportions, makeup style, skin tone, outfit, and overall appearance. White fitted dress and white flowing scarf throughout all frames. Light wind moves fabric naturally.
Photographed in a desert during sunset transitioning into blue hour. Professional DSLR photography aesthetic with direct on-camera flash. Flash is the dominant light source, creating a bright, sharply focused subject against a darker atmospheric desert background. Natural skin texture, realistic details, subtle film grain, authentic photography.
Top frame: close-up portrait. White translucent fabric softly draped near the shoulder and neckline. One hand resting near the collarbone. Looking into the camera. Wind gently lifting the fabric. Dramatic flash highlights, natural shadows, desert dunes in the background.
Middle frame: medium portrait. Structured white fashion top with gathered fabric and a circular metallic accent at the center. White scarf loosely wrapped around the neck and flowing in the wind. Body angled slightly sideways while facing the camera. Flash-lit subject against a warm desert sunset horizon.
Bottom frame: three-quarter portrait from a rear angle. White scarf wrapped around the head and shoulders. Subject turning toward the camera. Low sun near the horizon behind desert dunes. Flash freezes the subject while preserving the warm ambient atmosphere.
Editorial fashion photography, luxury magazine aesthetic, realistic proportions, high-contrast flash lighting, warm sunset tones, natural shadows, authentic DSLR image quality, shallow depth of field.
Preserve identity from the reference image. No change to facial structure, eyes, lips, nose, skin tone, makeup style, or body shape.
No text, no logos, no watermark, no CGI, no illustration, no stylization, no extra people.
