聊天GPT
電影般的夜間街頭肖像 — Same-to-Same
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：1,587 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
> ULTRA-REALISTIC 8K DSLR PHOTOGRAPH — SAME-TO-SAME REFERENCE RECREATION
Recreate the uploaded reference image with maximum photographic fidelity. Treat it as the absolute master blueprint and preserve the exact 9:16 composition, framing, crop, perspective, pose, expression, subject placement, wardrobe, lighting, colors, background, and depth of field.
Preserve the subject’s facial identity exactly—natural face shape, eyes, eyebrows, nose, lips, jawline, ears, beard, skin texture, hairline, pores, and natural asymmetry. No beautification, face redesign, or identity alteration.
Young man in a dark charcoal utility overshirt with rolled sleeves over a plain white T-shirt, standing naturally on a lively urban street at night. Warm and cool city lights, storefronts, pedestrians, and colorful practical lights create realistic cinematic bokeh. Keep the subject sharply focused with a softly blurred background.
Premium full-frame DSLR/mirrorless look, 50–85mm lens, f/1.8–f/2.2, eye-level perspective, realistic optical compression, natural skin tones, detailed hair and fabric texture, soft highlight bloom, subtle halation, controlled contrast, deep detailed blacks, restrained saturation, fine photographic grain, and authentic night-time lighting.
No CGI, illustration, plastic skin, artificial face, excessive HDR, fake bokeh, oversharpening, altered identity, changed hairstyle, changed clothing, changed pose, distorted anatomy, watermark, logo, or text.
If [MY FACE] is provided, replace only the subject’s face with [MY FACE] while preserving everything else exactly.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。
