聊天GPT
夜間情侶自拍-超寫實8K
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:2,239 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 RECREATION. Use the uploaded SCENE REFERENCE IMAGE as the absolute master blueprint. Recreate it with maximum photographic fidelity and zero creative deviation.
Preserve both subjects’ exact facial identities, anatomy, proportions, natural asymmetry, skin texture, hair, expressions, and distinctive features. The woman is foreground-left with a gentle head tilt and subtle closed-mouth smile, long tousled dark hair, red bindi, traditional metallic jhumka earrings, and a rust-orange patterned outfit with cream/brown motifs and striped border. The man stands slightly behind/right, naturally leaning toward her, with thick wavy dark hair, subtle facial hair, calm expression, and a loose white collared shirt with open upper buttons.
Maintain the exact pose, head positions, shoulder alignment, proximity, clothing folds, hair placement, framing, crop, and selfie perspective. Vertical ~4:5 composition, eye-level arm’s-length smartphone perspective with a natural 28–35mm equivalent look.
Recreate the nighttime urban setting with blurred streetlights, buildings, vehicles, trees, and city activity. Use warm overhead street lighting, mixed ambient tones, direct frontal flash, realistic facial shadows, natural skin highlights, and authentic nighttime bokeh.
High-end DSLR/modern smartphone photography, realistic flash exposure, natural dynamic range, detailed skin and hair microtexture, accurate fabric and jewelry, subtle lens bloom, mild low-light grain, restrained HDR, soft edge rendering, natural contrast, and warm cinematic color grading. Authentic candid photograph—not CGI, illustration, beauty retouching, or AI-generated-looking imagery.
NEGATIVE PROMPT:
Identity change, face swap artifacts, altered anatomy, beautification, skin smoothing, age change, ethnicity change, face morphing, altered hairstyle/expression/pose, composition changes, extra people/limbs, distorted jewelry or clothing, artificial bokeh, excessive HDR, oversharpening, waxy skin, CGI, 3D render, anime, cartoon, painterly effect, overexposure, crushed blacks, unnatural colors, beauty filter, watermark, text, logo, duplicate features, deformed eyes, distorted teeth, unrealistic hair or lighting.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。


