聊天GPT
眼睛說什麼字都說不出來
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:2,268 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
Create an ultra-photorealistic cinematic three-panel vertical portrait focused entirely on the subject’s eyes and upper face, using the reference image as the visual guide.
IDENTITY: Preserve the exact facial identity, eye shape, eyebrows, forehead, curly hairstyle, skin tone and natural facial proportions of the reference subject. Keep realistic pores, fine skin texture and tiny natural imperfections.
COMPOSITION: A vertical 9:16 image divided into three horizontal cinematic panels, each showing an extremely tight close-up of the same person's eyes and upper face. Maintain consistent facial features and framing across all three panels.
TOP PANEL: The subject looks directly toward the camera with an intense, serious and emotionally restrained expression. Both eyes clearly visible. Warm light illuminates the left side of the face while the right side fades into deep shadow.
MIDDLE PANEL: Same close-up and same person, but with a slightly more emotional expression. Eyes appear glossy with subtle moisture, creating a feeling of sadness and vulnerability without exaggeration.
BOTTOM PANEL: Strongest emotional expression. Subtle tears naturally forming beneath the eyes and a slightly reddened lower eyelid, conveying heartbreak and emotional exhaustion. Keep everything realistic and understated.
HAIR: Naturally curly black hair falling slightly over the forehead, with individually visible curls and realistic texture.
LIGHTING: Dramatic low-key cinematic lighting, warm golden/amber highlights, deep black shadows, strong contrast, soft falloff, subtle rim light around the hair. Dark minimalist background.
CAMERA: Extreme close-up portrait photography, 85mm macro-style lens, shallow depth of field, razor-sharp focus on the eyes, realistic catchlights, natural skin texture, subtle film grain, cinematic color grading, premium editorial photography, ultra-photorealistic, 8K detail.
MOOD: Quiet heartbreak, deep emotions, loneliness, emotional strength, introspection. Powerful but not exaggerated.
NEGATIVE: cartoon, illustration, artificial eyes, plastic skin, excessive tears, distorted face, different person between panels, inconsistent eyes, over-smoothing, beauty filter, unrealistic skin, extra facial features, blurry eyes, oversharpening.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。


