ペーパークラフトの世界に入りませんか?👼🪽
GPT Image 2 image prompt by のえる · original post
the prompt
concept:
- "a real human displayed like a living fashion doll inside a giant multi-layered shadow box"
- "the person remains fully photorealistic and human"
- "only the surrounding world is constructed from layered paper craft"
- "the entire art direction is derived from the reference image"
reference_direction:
identity:
- "faithfully preserve the reference persons recognizable facial identity"
- "preserve face shape, facial proportions, eyes, nose, lips, jawline, skin tone, distinctive asymmetry, age, and core hair silhouette"
- "no generic beautification or facial reinterpretation"
visual_analysis:
- "analyze the reference images dominant colors, accent colors, fashion, accessories, motifs, textures, objects, and overall mood"
- "derive the shadow-box palette and decorative world from those reference elements"
- "reinterpret meaningful reference details creatively rather than copying them literally"
- "small motifs may be enlarged, repeated, abstracted, or transformed into major environmental decorations"
- "do not introduce unrelated generic motifs"
world:
style:
- "extreme multi-layered paper-craft shadow box"
- "luxurious dollhouse display"
- "pop fantasy editorial"
- "dense reference-inspired decoration"
construction:
- "deep rectangular display frame"
- "many individually cut paper layers with visible gaps"
- "clear foreground, middle-ground, and background"
- "strong physical depth and distinct cast shadows"
- "foreground paper elements partially overlap the human"
- "avoid flat collage or printed-depth appearance"
subject:
type:
- "real human person"
- "photorealistic face and skin"
- "realistic human anatomy"
- "full body"
fashion:
- "redesign the outfit using the reference images colors, motifs, accessories, textures, and fashion identity"
- "evolve the reference styling into an elaborate editorial fashion look rather than copying the original outfit"
- "real wearable fabric, never paper clothing"
- "the outfit and shadow-box environment must share one coherent visual language"
pose:
- "sitting deep inside the shadow box"
- "cool fashion-editorial presence"
- "doll-like presentation while remaining naturally human"
- "legs extending forward and beyond the front edge of the box"
- "natural human proportions"
camera:
- "front-facing"
- "eye-level"
- "centered"
- "full-body"
- "35mm"
- "deep depth of field"
- "subject and paper layers sharply defined"
lighting:
- "studio lighting"
- "realistic light on human skin"
- "distinct shadows between paper layers"
- "lighting character adapted to the reference images mood"
critical:
must:
- "reference facial identity"
- "real human appearance"
- "reference-derived colors and decorative motifs"
- "extreme physical shadow-box depth"
- "legs breaking beyond the front frame"
avoid:
- "chibi or miniature person"
- "paper doll or paper-cutout person"
- "toy, plastic, or porcelain doll"
- "cartoon or illustrated face"
- "identity drift"
- "generic beautification"
- "flat collage"
- "paper clothing"
- "generic decorations unrelated to the reference"- model
- GPT Image 2
- type
- image
- shape
- 9:16
- shared
- September 17, 2026
- prompt
- 433 words
Made this? Fix the credit or remove it.




![Premium minimalist flat-vector travel poster for [CITY_NAME]](https://xviral.promptarium.app/gpt-image-2/700001.webp)




