Three-Person Merged Top-Down Selfie
This prompt merges three character photos into a high-angle Japanese/Korean-style selfie set in a movie theater, perfect for playful composites or virtual group photos.
The prompt
Fill in the blanks — 1 placeholder
Edit any of these and the prompt below updates. Copy gives you the filled-in version.
Do not swap faces, characters' faces must not change at all! Please merge the three people from Image 1, Image 2, and Image 3 into a three-person high-angle selfie. The composition is tight, the three subjects are very close, heads tilted slightly up, eyes looking directly at the lens, creating a strong visual impact. The person from Image 1 stands slightly forward in the middle, face similarity must be maintained; slightly hunching body, shooting angle is high-angle top-down, exaggerating head proportions, fitting typical Japanese/Korean visual selfie style. Simple and clean, further highlighting the character subjects. Visual style leans towards Japanese visual kei, overall image is high definition, shot with iPhone front camera, finally presenting an exquisite, stylish group photo effect. Requires seamless integration of characters into the scene, natural visual transition, overall lighting bright and even. Background is a super large cinema full of audience, two people with backs to the screen, screen displaying exquisite movie poster of "Zootopia 2" filling the screen.
Generate opens this prompt in the editor, ready to run.
Model
Seedream 4.5
Media
Image
Language
Chinese
Words
167
How to use this prompt
- 1Copy the prompt — fill in any placeholders first and the copy takes your version.
- 2Paste it into your image model, or press Generate to open it here with your own photo.
- 3Run it twice. Generation is not deterministic and the second pass is often the keeper.
- 4Finish it with the free tools: upscale, crop to the platform you need, compress before sharing.
Go beyond generating the image
Everything after the generation, on the same site.
More from Seedream 4.5
Same model, same dataset — Seedream 4.5 prompts people have shared.
All image categories
Use cases
Styles
- Photography (123)
- Cinematic / Film Still (138)
- Anime / Manga (44)
- Illustration (152)
- Sketch / Line Art (49)
- Comic / Graphic Novel (47)
- 3D Render (16)
- Chibi / Q-Style (4)
- Isometric (3)
- Pixel Art (2)
- Oil Painting (4)
- Watercolor (29)
- Ink / Chinese Style (13)
- Retro / Vintage (110)
- Cyberpunk / Sci-Fi (26)
- Minimalism (68)
Subjects
- Portrait / Selfie (164)
- Influencer / Model (6)
- Character (64)
- Group / Couple (23)
- Product (78)
- Food / Drink (48)
- Fashion Item (94)
- Animal / Creature (49)
- Vehicle (19)
- Architecture / Interior (55)
- Landscape / Nature (97)
- Cityscape / Street (62)
- Diagram / Chart (61)
- Text / Typography (119)
- Abstract / Background (283)