Japanese Street-Food Couple Date in Two Panels
Create a realistic nighttime street-food date photograph of a couple in a vertical, stacked two-panel composition, using reference images for their faces and the scene.
Example outputs
Prompt template
- Format
- Text Parameterized
- Inputs
- Reference_images (image) · Required watermark_text (text) · Optional · Default: mitralidia
- Outputs
- Two_panel_couple_photograph (image)
Create an ultra-realistic candid couple photograph in a vertical 9:16 composition split into exactly two horizontal panels stacked vertically. Use the reference photo for the poses, atmosphere, camera angles, and street-food moments.
TOP PANEL — MAN: Show a very handsome, masculine young man with a muscular, athletic physique, broad shoulders, and strong defined arms. He wears a fitted light baby-blue sleeveless T-shirt and a casual black beanie. He sits at a small outdoor Japanese street-food restaurant at night, holding wooden chopsticks with a delicious crispy golden fried snack and looking at the food with a playful, slightly serious expression. His other hand holds a small ceramic dipping bowl.
BOTTOM PANEL — WOMAN: Show a gorgeous, feminine, sexy, and adorable young woman wearing a fitted light baby-blue spaghetti-strap camisole with delicate white lace trim, matching the original reference outfit, paired with casual light-wash jeans. Her long dark hair is styled in two cute braids, with a few loose strands framing her face. She wears subtle earrings and natural Korean-style makeup with glowing skin, soft pink blush, and glossy nude-pink lips.
She sits on a wooden outdoor chair, playfully tilting her head backward while making a cute kissing expression and holding wooden chopsticks near her face. Place a plate of crispy golden fried street food and small ceramic dipping bowls on the table in front of her.
ENVIRONMENT AND PHOTOGRAPHY: Create an authentic Japanese nighttime street-food atmosphere, with illuminated LAWSON convenience-store signage in the background, an urban Japanese sidewalk, passing pedestrians, parked cars, wooden restaurant chairs, and warm street lighting mixed with cool blue neon lights. Include beautiful, natural background bokeh.
PHOTOGRAPHIC STYLE: Ultra-realistic candid smartphone photography, authentic spontaneous expressions, natural skin texture, realistic body proportions, sharp facial details, warm cinematic night colors, balanced exposure, realistic shadows, subtle photographic grain, high clarity, premium lifestyle photography, 4K quality.
IDENTITY: Use both uploaded reference faces as the exact identity references. Preserve 100% of their original facial identities, facial structures, eyes, noses, lips, jawlines, and natural expressions. Do not beautify or redesign their faces. No face swapping, identity drift, or facial alterations.
COMPOSITION: Exactly two horizontal panels stacked vertically, matching the original reference composition. The top panel features the man eating street food; the bottom panel features the woman posing playfully. Both panels must look like photographs taken during the same date, at the same location and time.
Add transparent watermark text reading ${watermark-text:mitralidia}.
Do not add extra panels, illustration, cartoon styling, distorted anatomy, extra fingers, blurry faces, artificial-looking skin, outfit color changes, or duplicated people.
Replace the ${…} placeholders with your values
before using this prompt. See parameters
and defaults below
Parameters
Replace each placeholder with your own value, or use its default when one is provided
${watermark-text:mitralidia}
watermark-text
Default: mitralidia
What you'll create
The upper panel shows a man with chopsticks and a small dipping bowl, while the lower captures the woman playfully. Matching location, time and photographic treatment tie the stacked moments together.
Usage instructions
Frequently asked questions
Can extra panels be added?
No. Exactly two horizontal panels are stacked vertically.
Does the man look at the camera?
He looks toward the food in his chopsticks.
Comments
Tried this prompt? Share what you made, compare results, or ask a question