
Original document
GPT Image 2 · en
Scene:
A square frame organized around a ruthless one-point perspective, with the vanishing point pinned dead center and every wall seam, ceiling light, floor tile, and doorway converging toward it. The palette should lean cold white, muted red, steel gray, and hospital green, with a monumental sense of emptiness and institutional order. Build a carpeted corporate passage linking meeting rooms, with linear wall sconces, wood veneer panels, and narrow red carpeting aimed directly at the center.
Subject:
Place one small human figure or one important object along the central axis so the architecture dominates the scale relationship. The corridor or interior must feel long, controlled, and mildly oppressive, as if the camera is witnessing a space designed by precision rather than comfort. A single rolling briefcase stands on the axis under a ceiling vent, as if its owner just vanished beyond the frame.
Important details:
The essential grammar is exact perspective discipline, polished surfaces, measured repetition, and low emotional temperature. Preserve straight lines, clean threshold spacing, and a believable practical-lighting logic. The image should feel clinical, deliberate, and spatially hypnotic rather than decorative or chaotic. Hold the palette to beige, smoke gray, and disciplined oxblood so the corridor feels expensive but emotionally sterile.
Use case:
Single-Point Perspective Clinical Corridor single study 6/9.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, no fisheye distortion, no crooked horizon, no cluttered props, no graffiti, no colorful signage, no broken anatomy, no low-detail background, no accidental text fragments.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.





