聊天GPT
電影級低角度肖像 — 最大攝影水平
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：1,279 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
Recreate the uploaded reference image with maximum photographic fidelity. Preserve the subject’s exact facial identity, natural features, hair, beard, skin texture, expression, pose, clothing, composition, camera angle, and proportions without beautification or alteration.
Use a vertical 3:4 close-up, captured from a noticeably low angle with a 35–50mm DSLR/mirrorless look at f/1.8–f/2.8. Preserve the dark architectural surroundings and geometric window structure. Strong warm sunlight enters from the upper-right, creating realistic highlights and deep directional shadows across the face, hair, and charcoal crew-neck T-shirt.
Maintain natural skin pores, individual hair and beard strands, fabric texture, realistic depth of field, subtle lens falloff, authentic dynamic range, restrained cinematic color grading, and natural film grain. The final result must look like an untouched high-end DSLR photograph, not CGI or AI-generated.
Negative:
AI look, CGI, plastic skin, beauty filter, face morphing, altered identity, facial reconstruction, changed pose/expression, changed clothing/background, excessive bokeh, fake HDR, oversaturation, crushed blacks, blown highlights, artificial lighting, excessive sharpening, distorted anatomy, artifacts, text, logo, watermark.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。
