Bilingual Retro Naive Editorial Poster

Bilingual Retro Naive Editorial Poster

Use the system imagegen skill and image-generation tool. Treat each uploaded photograph as the edit source. Build a direct but imaginative dialogue between the real upper image and a lighter, stranger, hand-made lower interpretation.

Output contract

• Produce one independent raster poster for every uploaded photo. Never combine photos into a collage, contact sheet, comparison layout, diptych, or shared canvas.
• Use a 3:4 portrait canvas with one exact horizontal 50/50 division. Keep truthful photography in the upper half and the retro-naive illustration in the lower half. Do not let either medium invade the other.
• Preserve immediate correspondence through subject identity, pose, emotion, direction, relationship, defining color, and one or two symbolic features.
• Use the user's preferred or current language for primary type. Add the other language only as a small optional editorial echo when useful.
• Keep text optional and sparse.
• Return every poster as a separate image. Keep user-facing commentary to one short status sentence only when necessary.

Preserve the photographic upper half

Keep the source recognizably photographic. Preserve subject identity and count, anatomy or construction, proportions, emotion, pose, silhouette, direction, spatial relationships, real materials, natural light, and the original color atmosphere.

Apply only restrained magazine- or exhibition-photography color correction. Do not redraw, vectorize, beautify, materially relight, or apply illustration, typography, paper texture, or graphic overlays to this region.

Crop deliberately for the upper half. When more area is needed, extend only plausible sky, ground, wall, water, foliage, or environmental background. Never stretch, warp, mirror, duplicate, or redesign the subject.

Reduce the source to a playful truth

Before prompting, privately identify:

1. the one hero subject or inseparable relationship carrying recognition;
2. its decisive posture, emotion, silhouette, proportion, action, direction, and three to six identity cues;
3. the source's gentle humor, awkwardness, tenderness, romance, attitude, or quiet eccentricity;
4. one feature that may be slightly enlarged, tilted, simplified, repeated, or symbolically transformed;
5. the minimum environment line or geometric support needed for place and story;
6. the source's high-key color families and one possible complement accent;
7. whether text is useful, and if so, the exact primary-language title and optional secondary-language echo.

Do not expose this analysis or repeat the user's design brief.

Rebuild with concise handmade form

Use simplified masses, slightly exaggerated proportion, symbolic features, and a few thin hand-drawn contours. Preserve the source's main pose and relational space.

Let one local feature carry personality: a hat, posture, branch, wheel, roof, garment, animal expression, vessel opening, path, shadow, or another memorable source cue.

Use moderate asymmetry and minor hand error. Keep forms concise, light, and editorial. Do not turn every subject into a person or mascot; architecture, objects, vehicles, plants, and landscapes retain their own functional identity.

Avoid chibi anatomy, anime faces, emoji expressions, generic cute animals, sticker borders, smooth vector symmetry, detailed realism, and exaggeration that changes identity.

Use one source-backed visual metaphor

When the photograph supports it, translate one relation into a simple humorous or poetic visual move: a path becoming a ribbon, a plant leaning like a gesture, a window holding a face-like opening, a garment becoming a sheltering shape, or another source-specific idea.

Keep the metaphor secondary to recognition. It should arise from action, pose, shape, function, or relation—not from decorative whim.

If no metaphor improves the image, use straightforward simplified observation instead. Avoid forced puns, random surreal props, repeated symbols, and generic hearts, stars, suns, arrows, or speech bubbles.

Compose with modernist breathing room

Use one clear center of gravity, broad off-white or pale blank areas, one or two simple geometric supports, and a restrained reading path.

Place the hero according to source posture and direction. It may be centered, lightly offset, cropped, or extended. Use scale contrast, diagonal/axial rhythm, overlap, and asymmetry to create balance.

Add only a few hand-drawn lines, geometric symbols, pale washes, or abstract forms that support the hero. Avoid dense Bauhaus decoration, multiple focal points, complete background scenes, patterned borders, equal object rows, and reusable template grids.

Present the result as a flat illustration page. Do not add a frame, wall, book mockup, desk, gallery room, fashion presentation board, product card, or e-commerce context.

Build the high-key limited palette

Derive the lower palette from the upper photograph. Use high lightness, low-to-medium saturation, and neighboring hues to unify the atmosphere. Add one small complementary jump when it strengthens a focal detail.

Use off-white or a pale source hue as the largest area and roughly three to six active color families including line and background. Concentrate stronger color on the hero and one support element.

Keep the palette gentle, stylish, lightly nostalgic, and clear. Avoid fixed retro palettes, dark grounds, muddy brown, overall grayness, neon, rainbow color, candy overload, smooth gradients, and equal-area multicolor.

Assign each physical medium a role

Use a unified mixed-media system:

• watercolor for translucent air, shadow hint, or receding field;
• gouache for clear flat hero masses;
• colored pencil for fine local structure and material cues;
• oil pastel for soft tactile emphasis or irregular focal edges;
• dry brush for motion, hair, foliage, ground, worn transition, or print-like friction;
• thin ink/pencil for only the essential contour and gesture.

Show visible paper grain, exposed paper, broken strokes, rough edges, imperfect brushwork, light dirty color at overlaps, and small registration drift.

Avoid smooth vector fill, photorealistic lighting, thick oil paint, one medium used everywhere, watercolor flooding, crayon-heavy child drawing, polished digital brushes, 3D form, plastic gloss, and texture applied uniformly.

Integrate optional language-aware type

When text improves the work, derive one short title, keyword, subject/place name, number, or phrase from the source. Keep primary wording in the user's language.

The title may use slightly naive retro hand lettering. Tiny secondary text may use a gentle sans serif or serif. A small translation or alternate-language word may appear as an editorial echo, never as a competing duplicate title.

Allow slight tilt, stretch, offset, overprint, or registration drift. Align type to the hero gesture, contour, wash, geometric support, horizon, or blank area.

Use a year, place, or number only when supplied or reliably known. Avoid false facts, dense copy, large commercial headlines, multiple fonts, advertising slogans, logos, brands, signatures, pseudo-language, or watermarks.

Compile the image-edit prompt

Write one source-specific production prompt per input photo containing:

1. one independent 3:4 portrait output and the exact 50% photography / 50% retro-naive split;
2. the upper-half preservation invariants, restrained magazine/exhibition color correction, crop, and any safe background extension;
3. the lower hero, emotion, pose, direction, relationship, three to six identity cues, and source attitude/story;
4. the exact simplification, local exaggeration, symbolic feature, optional source-backed metaphor, and deleted detail;
5. the hero forms, thin imperfect contours, center of gravity, asymmetry, crop/extension, geometric supports, reading path, and blank-space shape;
6. the exact high-key neighboring palette, pale ground, hero/support colors, and small complement accent;
7. the exact watercolor, gouache, pencil, oil-pastel, dry-brush, and line roles, plus paper grain, exposed paper, rough edges, broken marks, overlap color, and registration drift;
8. the optional exact primary-language title/keyword/name/number/phrase and optional small secondary-language echo, with factual support and gesture/contour/wash/support/horizon/blank placement;
9. hard exclusions: no multi-photo layout, no altered or redrawn upper subject, no approximate split, no literal detail tracing, no generic cute mascot, no forced visual pun, no smooth vector, no complex realistic scene, no dense Bauhaus decoration, no dark/muddy palette, no polished digital rendering, no 3D, no e-commerce look, no false year/place/number, no large ad title, no logo, and no watermark.

Use concrete source attitude, form, exaggeration, metaphor, color, medium job, and language hierarchy rather than relying on adjectives such as “premium,” “retro,” “naive,” “fashionable,” or “playful.”

Inspect and correct

Review every result at full size and thumbnail size. Verify the exact split, truthful upper photography, recognizable lower subject and emotion, concise handmade form, restrained source-backed metaphor, large breathing room, high-key limited palette, purposeful mixed media, and optional language-aware type.

If it looks like a generic cartoon, restore source pose, emotion, structure, and color memory. If it is overly polished, introduce restrained hand errors and broken media. If the metaphor hides identity, remove it. If the page becomes busy, delete support shapes and text. If bilingual text competes, keep only the primary language or reduce the echo. Regenerate with one targeted correction when needed.