聊天GPT
肖像 — 電影 — 兩次之內
為聊天GPT 編寫。不同型號的結果和參考照片支援可能有所不同。
使用提示詞前
列出的工具:聊天GPT。完整說明:1,451 個字元。 1 個提供的範例影像。
This prompt mentions reference images. Read its instructions for the required number of photos and identity constraints. We have not independently tested this prompt. Tool labels and supplied images are not a guarantee of compatibility or output quality.
完整提示詞
Use the uploaded image strictly as the identity source and create a vertical 9:16 cinematic portrait featuring the same person twice within one composition.
Accurately retain the subject’s recognizable face, facial structure, natural skin tone, hairstyle, and overall appearance from the reference.
Build the scene with a large, softly enlarged portrait of the subject occupying the background. Keep this oversized version clearly recognizable but atmospheric, using a warm faded-sepia appearance, gentle desaturation, low contrast, and soft focus so it blends naturally into the dark background.
In front of it, place a smaller full-color version of the same person toward the lower center-right. This foreground subject should remain sharp and realistic, keeping the original clothing and natural appearance from the reference, with a relaxed posture and subtle soft smile.
Separate the foreground figure from the background using a delicate warm golden glow along the hair and shoulders. Use restrained cinematic lighting, warm highlights, deep surrounding shadows, and shallow depth of field. The large portrait should feel distant and dreamy while the smaller foreground figure remains the visual focus.
Create a refined social-media-style photographic finish with natural skin texture, believable anatomy, detailed hair, smooth light transitions, and realistic depth. No text, logos, stickers, interface elements, signatures, or watermarks.複製完整的提示並根據要求使用您自己的參考照片。
此提示提供的範例影像。此提示尚未經過我們網站的測試。


