091写真を金属レリーフに

Generate this image now. Draw it directly.
出力サイズは1080×1440ピクセル、3:4の縦型SNS投稿にしてください。
Create exactly one finished rectangular framed metal bas-relief portrait, filling the canvas edge to edge, default 3:4. One premium sculptural illustration with authentic chased metal, deep relief planes, fine chisel marks and dark recessed shadows. The narrow integral rectangular frame is part of the artwork. No room, wall, product mockup, collage, source inset, circular outline, coin or medal.
PHOTO RULE: Use the attached photo as identity reference. Preserve face shape, eye spacing, nose, jaw, hairstyle, expression, age and head direction; for a pet preserve species, ear asymmetry, muzzle, fur silhouette and markings as relief boundaries. Keep recognizable identity, translating all skin, eyes, hair, fur, clothing and background into the selected metal. Do not leave natural-colored skin or photographic patches. If multiple subjects appear, portray only the foremost subject. "【写真】" is an identification note, never visible text. If no photo is attached, invent an original adult or pet matching the note, or a fluffy asymmetrical-eared dog when the note is empty. Never stop to request an upload.
"【金属】" chooses one metal only: 金 (gold), 銀 (silver), 銅 (copper). Empty or unrecognized metal defaults to 金. Gold has warm luminous highlights and dark gold recesses; silver has cool white highlights and graphite recesses; copper has warm reddish highlights and oxidized brown recesses. Keep the same deep sculptural treatment across choices. Adapt restrained carved botanical accents at lower corners to the subject without covering the face. Use raking light to reveal relief, not plastic shine. Preserve the subject's mood: gentle daylight for a smiling portrait, subdued cool light for a contemplative portrait, warm dramatic side light for a pet.
Render no text, names, letters, numbers, dates, labels or inscriptions, even if present in the photo. No logos, trademarks, branded accessories, existing characters, signatures, watermarks, recognizable famous likenesses or imitation of named artists or artworks. Anatomically coherent, no extra eyes, ears or limbs. Deliver one pure finished image, not a manufacturing drawing, height map or 3D model.

[Final output format]
出力サイズは1080×1440ピクセル、3:4の縦型SNS投稿にしてください。
This selected output format takes priority over any fixed canvas size or aspect ratio elsewhere in this template. Adapt the composition, text zones and margins to the selected canvas without stretching or cropping essential content. Device safe areas apply only when a device is selected. Never print these format instructions on the image.