Affiche sérigraphie gestalt scindée avec silhouette Morandi

Turn each photo I upload into an independent, high-end design poster; no multi-image collage, each photo output separately. Use an overall 3:4 vertical composition, with the top and bottom areas strictly 1:1 in height, each occupying 50% of the frame. The top half keeps the original photo, preserving the subject's structure, real texture, natural light and shadow and original color atmosphere, with only slight high-end photographic color grading, giving it the quality of art magazines, independent publications and exhibition photography. To fit the frame, the sky, ground or environmental background may be naturally extended, but the subject must not be stretched, distorted or altered. The bottom half extracts the most recognizable subject, contour, pose and narrative relationship from the photo, reconstructing them into a minimalist gestalt positive-negative silkscreen print. Keep the subject's most recognizable outer contour, while distilling from the original image one most closely related environment, structure, object or symbolic imagery, letting the two merge into a single complete figure through contour interlocking, negative-form cutouts, shared boundaries and figure-ground reversal: from afar you first recognize the subject, up close you discover another layer of scene or imagery hidden within. Do not simply stack the two objects, but let the edge of one shape naturally become the boundary of the other. The composition keeps a single visual anchor, a clear silhouette and large areas of white space, determining the position according to the subject's own pose, orientation, center of gravity and contour flow; it may use off-center placement, partial cropping, vertical nesting or edge alignment, forming a clever visual transition between positive and negative form. Auxiliary elements must be minimal, letting the visual interest come mainly from the instant switch between "what you see" and "what you see again", rather than complex decoration. The colors use a healing, soft and high-end Morandi pastel system, referencing the low-saturation color logic of dreamlike architectural spaces: misty pink, soft apricot, gray green, sage, pale terracotta, cream yellow, misty blue, gray purple, etc., harmonizing with each other, on a light-colored paper ground, selecting the most spirited color relationships from the original photo to soften and recompose. Keep it to 2–4 colors overall, building layers through color temperature, brightness difference and area ratio; colors clean, soft and airy, avoiding dirty gray, heavy aging, fluorescent colors and high-contrast commercial palettes. Retain slight paper fibers, silkscreen grain, ink coverage and registration deviations, giving the figure a real printmaking touch, but keep edges sharp and color blocks flat, without gradients, 3D volume, realistic shadows or smooth CG texture. Text also participates in the positive-negative form composition. Distill a short English title from the photo's identity, place, action, emotion, state or symbolic meaning, paired with a minimal amount of place, numbering, state words or tiny annotations. The text can be arranged along the subject's contour, the negative-form boundary, a horizontal baseline or the white-space axis, and can also be partially embedded in the figure, cut off by the contour or occluded by color blocks, making the text feel like part of the print structure rather than a title pasted on afterward. Keep the whole thing small, restrained, clever and with a high-end editorial feel. The whole presents a visual character combining minimalist geometry, gestalt optical illusion, soft Morandi colors, silkscreen printmaking and poetic narrative. Whether the subject is a person, animal, plant, building, object, vehicle or natural landscape, it should keep the dual-reading experience of "recognize the subject at a glance, discover the hidden imagery at second glance", avoiding multiple parallel focuses, direct collage, complex illustration, cartoon feel, commercial promotional feel and template feel. Text preference: english