聊天GPT
肖像中的肖像 — 椅子一條腿
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:1,423 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 as the PRIMARY VISUAL REFERENCE. Preserve the exact recognizable facial identity, facial structure, hairstyle, skin tone and natural body proportions of the subject. Recreate the same sophisticated indoor composition with the subject seated casually on a wooden chair, one leg crossed over the other, holding a smartphone naturally in one hand. Preserve the white-and-red gradient shirt, black trousers, white sneakers and wristwatch. Behind the subject, create a large framed black-and-white artistic portrait of the same person, creating a striking portrait-within-a-portrait effect. Keep the minimal warm interior, light neutral wall, wooden frames and small decorative objects on the floor. Maintain the soft natural indoor lighting and shallow depth of field. The seated subject MUST look directly into the camera, while the framed portrait can retain a relaxed side-facing expression. Photorealistic editorial lifestyle photography, realistic skin and fabric texture, natural colours, 1122 × 1402 px, vertical.
Negative Prompt
Face distortion, wrong identity, pasted face, inconsistent identity between subject and framed portrait, altered hairstyle, bad anatomy, distorted hands, extra fingers, warped smartphone, distorted chair, incorrect clothing, unrealistic framed portrait, artificial skin, excessive saturation, CGI look, blur, unnatural shadows, random text, watermark, AI artifacts.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。


