OpenZhenZhen AI プロンプトギャラリー · gpt-image-2.5 · image

無限再帰の中の華麗な額縁肖像

Ornate Framed Portrait in Infinite Recursion

「無限再帰の中の華麗な額縁肖像」は gpt-image-2.5 を使用した作品です。OpenZhenZhen で元のプロンプトと生成例を確認し、創作の参考にできます。

無限再帰の中の華麗な額縁肖像 — AI image creation
無限再帰の中の華麗な額縁肖像

プロンプト

中国語プロンプト

创作一幅超现实写实风格的视觉递归肖像,主体为一位明确成年的人物,身份绑定至已配置参考ref-10a-v1。主画面和每一层嵌套图像都保持同一人物可辨识的面部身份及自然比例。人物双眼睁大、神情惊讶,朝镜头举着一个华丽的大相框。相框内是一张照片,照片里的仍是同一人物,手持同一相框并保持相同惊讶表情;如此连续重复多层,并逐渐缩小,形成向中心螺旋收束的套框幻觉。主画面人物微微向前倾,仿佛被轻柔吸入框中图像,但姿态停留在悬置瞬间,不表现撞击、受伤或超出惊讶表情的痛苦。让递归结构一目了然:每层内图都重复人物、画框和表情,只是尺度更小,形成容器与内容彼此镜映的奇异循环。使用干净克制的室内背景、受控透视和柔和电影光;最大相框清晰锐利,内层重复逐渐收细,但不混乱。整体超现实而连贯,不增加人物、文字、标志或水印。“无穷”以有限数量的嵌套画框表现为视觉幻觉,不声称真实无限细节或精确分辨率。

英語プロンプト

Create a surreal, photorealistic visual-recursion portrait of one clearly adult person bound to configured identity reference ref-10a-v1. Preserve the same recognizable face and natural proportions in the main subject and every nested image. The person has wide, startled eyes and holds a large ornate picture frame toward the viewer. Inside the frame is a photographic image of the same person holding the same frame with the same surprised expression; repeat this exact frame-within-frame scene through several progressively smaller nested iterations, spiraling toward a tiny central vanishing point. Show the main person leaning forward as if being gently drawn into the framed image, but keep the pose suspended and non-injurious, with no impact or distress beyond the startled expression. Compose the recursive frames so the loop is immediately readable: each inner image repeats the person, frame and expression at smaller scale, producing a strange visual loop in which container and content mirror one another. Use a clean subdued interior, controlled perspective and soft cinematic light; keep the largest frame sharp and let the nested repetitions taper into fine detail without becoming chaotic. Surreal but coherent, no extra people, text, logos or watermark. Treat “infinity” as a visual illusion represented by finite nested frames; do not claim literal infinite detail or exact resolution.

API モデルを見る

画像・動画プロンプトをもっと見る · 画像を生成 · 動画を生成 · AI キャンバスで作成