提示詞/GPT Image 提示詞/提示詞詳細資料圖片提示詞一致 3x3 多角度場景表生成寫實 3x3 網格，從多個連貫的相機角度呈現相同的參考主體與環境。此提示詞使用了參考圖片立即使用讚1,207瀏覽388,320收藏940儲存此提示詞或分享給朋友。分享提示詞提示詞複製Using the provided reference image, create a coherent multi-view scene reconstruction with the same person and retail warehouse environment. Generate a {argument name="aspect ratio" default="16:9"} image divided into an exact {argument name="grid layout" default="3x3"} contact sheet. In all 9 panels, keep the subject’s identity, clothing, sunglasses, display table, and store setting consistent, but redraw the surrounding scene from varied camera angles and distances so the person’s standing position and environment remain spatially plausible. Required 9 views: 1. Front-facing medium view from across the sunglasses table. 2. Rear view showing the subject’s back and the table ahead. 3. Right-side profile view beside the display. 4. Left-side profile view from closer to the table. 5. Three-quarter front view similar to the reference. 6. Three-quarter front/right view with shelves behind. 7. Rear-left view looking over the table. 8. Rear three-quarter view with the subject centered near the display. 9. Wide elevated view showing the full sunglasses table, subject, and surrounding warehouse aisles. Use {argument name="model or style note" default="GPT-image-2.5 Sunburst"} style photorealistic synthesis. Emphasize front, back, left, and right views as mandatory. Do not add captions, speech bubbles, borders beyond the clean white grid dividers, or any extra people.展開提示詞
