Skip to content
ImagesFree

Conceptual Silhouette Void Split Comparison

Generates a vertical 3:4 split composition pairing an intact photograph on top with a minimalist, charcoal-black silhouette and warm-white negative space on the bottom.

OpenAI

Source previews

These example images come from the original source. DROPIDEA has not independently tested these examples; results depend on the model and settings.

Ready to personalize

Your prompt

English

Replace the red variables with your information. Copy the full prompt and use it with your preferred agent.

TXT MD English version
Create a vertical 3:4 before/after transformation image, with the style named 【CONCEPTUAL SILHOUETTE VOID】. Applicable to people, animals, plants, architecture, cities, vehicles, food, objects, interiors, nature, and travel photos — not limited to "journey" or any single scene. The upper portion (roughly 42%–48%) preserves the photo as is: accurately maintain the main subject, count, pose, viewpoint, spatial relationship, lighting, material, and recognizable colors; only apply natural, publication-grade color grading. Do not illustrate it, change the location, or add/remove key subjects. Leave a clean warm-white seam between top and bottom; do not add thick borders, photo labels, or original/AI corner marks. The lower portion should not replicate the entire photo, nor draw a complete illustration. First distill the single most meaningful semantic relationship, action, silhouette, or material rhythm from the original photo: for example, a figure's standing pose, an animal's tail arc, a flower opening, a branch's extension, a road's bend, an architectural arch line, a vehicle's speed direction, the circular relationship of bowls and plates, the repeated rhythm of windows, the undulation of a ridgeline, or ripples on water. Select only one main vehicle, and reconstruct it into a flat charcoal-black, ink-black, or deep graphite abstract silhouette, banded contour, block symbol, or single linear gesture. The shape must be traceable back to the real subject in the upper portion, but may be elongated, reduced, offset, rotated, cut off, or extended across the negative space. Maintain 60%–80% continuous warm-white negative space in the lower portion; the paper surface should have only extremely subtle fibers and matte grain. The main shape's edges may have slight halftone dots, dry-brush grain, or uneven ink density, but overall it must be restrained, flat, and clean — not bleeding into ink-wash, nor becoming a refined line drawing. People and animals may be compressed into faceless, untextured-clothing overall silhouettes; architecture, plants, food, and objects retain only the minimum structure that identifies the shape relationship, without restoring photo detail. Use at most one or two extremely thin auxiliary lines to suggest direction; do not add a second abstract object. The abstract shape is primarily charcoal-black, ink-black, or deep graphite, with at most one extremely low-saturation accent color extracted from the original image; the color varies with the photo, and does not apply a fixed blue, green, or city palette. A single line of small, precise English editorial caption may be added at the bottom, describing the semantics, e.g., ONE GESTURE / OPEN SPACE, ARC / STILLNESS, or SINGLE FORM; do not add a large headline, long paragraph, or fabricated location information. Overall, the upper portion should carry photographic fact; the lower portion should carry a single concept, shape judgment, and negative-space rhythm. Prohibited: full-bleed scene redrawing, multiple floating objects, continuous route nodes, complete city skyline templates, photo windows, full-page ink-wash, colored geometric posters, heavy paper-craft, 3D rendering, automatic outlining, logos, watermarks, fake Chinese, AI corner marks, and original-image corner marks.
Preview results

Your prompt

Create a vertical 3:4 before/after transformation image, with the style named 【CONCEPTUAL SILHOUETTE VOID】. Applicable to people, animals, plants, architecture, cities, vehicles, food, objects, interiors, nature, and travel photos — not limited to "journey" or any single scene. The upper portion (roughly 42%–48%) preserves the photo as is: accurately maintain the main subject, count, pose, viewpoint, spatial relationship, lighting, material, and recognizable colors; only apply natural, publication-grade color grading. Do not illustrate it, change the location, or add/remove key subjects. Leave a clean warm-white seam between top and bottom; do not add thick borders, photo labels, or original/AI corner marks. The lower portion should not replicate the entire photo, nor draw a complete illustration. First distill the single most meaningful semantic relationship, action, silhouette, or material rhythm from the original photo: for example, a figure's standing pose, an animal's tail arc, a flower opening, a branch's extension, a road's bend, an architectural arch line, a vehicle's speed direction, the circular relationship of bowls and plates, the repeated rhythm of windows, the undulation of a ridgeline, or ripples on water. Select only one main vehicle, and reconstruct it into a flat charcoal-black, ink-black, or deep graphite abstract silhouette, banded contour, block symbol, or single linear gesture. The shape must be traceable back to the real subject in the upper portion, but may be elongated, reduced, offset, rotated, cut off, or extended across the negative space. Maintain 60%–80% continuous warm-white negative space in the lower portion; the paper surface should have only extremely subtle fibers and matte grain. The main shape's edges may have slight halftone dots, dry-brush grain, or uneven ink density, but overall it must be restrained, flat, and clean — not bleeding into ink-wash, nor becoming a refined line drawing. People and animals may be compressed into faceless, untextured-clothing overall silhouettes; architecture, plants, food, and objects retain only the minimum structure that identifies the shape relationship, without restoring photo detail. Use at most one or two extremely thin auxiliary lines to suggest direction; do not add a second abstract object. The abstract shape is primarily charcoal-black, ink-black, or deep graphite, with at most one extremely low-saturation accent color extracted from the original image; the color varies with the photo, and does not apply a fixed blue, green, or city palette. A single line of small, precise English editorial caption may be added at the bottom, describing the semantics, e.g., ONE GESTURE / OPEN SPACE, ARC / STILLNESS, or SINGLE FORM; do not add a large headline, long paragraph, or fabricated location information. Overall, the upper portion should carry photographic fact; the lower portion should carry a single concept, shape judgment, and negative-space rhythm. Prohibited: full-bleed scene redrawing, multiple floating objects, continuous route nodes, complete city skyline templates, photo windows, full-page ink-wash, colored geometric posters, heavy paper-craft, 3D rendering, automatic outlining, logos, watermarks, fake Chinese, AI corner marks, and original-image corner marks.

Variable guide

ONE GESTURE / OPEN SPACE

Brief English editorial caption summarizing the semantic visual concept

The editorial caption example inside the penultimate paragraph describing the lower section

Example value

ONE GESTURE / OPEN SPACE

What you get

  • A dual-tier 3:4 vertical composition balanced with a subtle warm-white seam.
  • Upper half presenting the original photograph with faithful subjects, colors, and publication-grade grading.
  • Lower half showcasing a single abstract silhouette in ink/charcoal surrounded by 60%–80% textured warm-white void.
  • A minimal, publication-style English caption that ties the conceptual form together.

Use cases

  • Creative editorial posters and portfolio displays contrasting reality with abstraction.
  • Travel and architectural photo series presented with sophisticated gallery-level branding.
  • Minimalist social media carousels and visual identity assets.

How to use

  • Upload or provide a high-quality reference photo (portrait, architecture, nature, street scene, or object).
  • Submit the prompt alongside your reference image to instruct the model on the 3:4 vertical split composition.
  • Optionally customize the editorial caption text (e.g., replace 'ONE GESTURE / OPEN SPACE' with 'ARC / STILLNESS' or 'SINGLE FORM') to match your image concept.
  • Review the resulting output to ensure the top holds photographic fact while the bottom distills a single minimalist silhouette into spacious negative space.

Related prompts