聊天GPT

肖像 — 特寫、自拍 — 唯一來源身份

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:5,374 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 user photo as the **ONLY source of identity** for the subject. Preserve the user's exact recognizable facial identity, facial structure, proportions, face shape, forehead, eyebrows, natural eye shape, nose, lips, jawline, cheeks, chin, skin tone, natural skin texture, hairline, and other distinctive features. Do not copy the identity or facial appearance of the reference subject. Do not beautify, reshape, slim, age, de-age, masculinize, feminize, or reinterpret the user's face. The user's original photo provides WHO the person is; the reference image provides HOW the person looks, poses, expresses emotion, and interacts with the scene. Recreate the reference as a vertical three-panel photographic collage consisting of three horizontally stacked portrait frames separated by thin clean white divider lines, with all three panels featuring the same user identity in the same warm indoor environment. Maintain the overall 9:16 vertical composition and realistic high-end smartphone/editorial photography aesthetic. PANEL 1: Create a close-up selfie-style portrait from approximately chest/shoulder level upward. The subject is positioned slightly right of center, with the head gently tilted and turned toward the camera, looking directly into the lens. Reproduce the reference's playful expression: softly puckered lips in a subtle kiss/duck-face expression, relaxed eyes, naturally raised/relaxed eyebrows, and a playful but calm mood. Long, voluminous dark hair falls naturally over the subject's left side and shoulder with loose waves and realistic individual strands. The subject wears a deep burgundy/maroon long-sleeved top with a wide/open neckline. One arm extends naturally toward the camera as if holding a phone for a selfie. Add delicate hand-drawn white doodle outlines around the silhouette, including small curved accent marks and a simple heart on the right side, as seen in the reference. PANEL 2: Recreate the same indoor setting, lighting, wardrobe, hairstyle, and identity, but change the pose and expression exactly as shown: the subject faces the camera with the same playful puckered-lips expression while raising one hand beside the face and holding a small bright red/orange flower directly over one eye. The fingers delicately pinch the flower stem/head, with the flower centered over the eye and partially covering it. Keep the other eye visible and directed toward the camera. Preserve the natural head angle and relaxed shoulders. Add subtle white hand-drawn doodle strokes around the hair and upper body. Keep the burgundy long-sleeved top and realistic hair texture unchanged. PANEL 3: Recreate the same person and environment again, but with a softer, friendly expression: the subject tilts the head gently toward one shoulder, looks directly into the camera, and gives a natural closed-mouth smile. The eyes should appear warm and relaxed, with a gentle approachable expression. Hair remains long, dark, voluminous, and naturally wavy, cascading over one shoulder. Keep the same burgundy/maroon top and neckline. Add playful hand-drawn doodles around the silhouette, including small sparkle/flower-like symbols on the left and a loose white outline around portions of the hair and shoulders. ENVIRONMENT: All three panels take place in the same cozy sunlit interior with warm beige/taupe walls. A leafy green houseplant is visible in the left background, softly blurred. On the right side, strong late-afternoon sunlight enters through a window or architectural opening, creating distinctive geometric rectangular window-frame shadows across the warm wall. Maintain the same background arrangement in every panel. Use shallow depth of field so the subject is sharply detailed while the background remains softly blurred but recognizable. LIGHTING: warm golden-hour sunlight from the right side of the frame, producing soft directional highlights across the face, hair, neck, and clothing, with gentle natural shadows and a warm amber color temperature. Preserve realistic skin texture, subtle facial highlights, natural pores, realistic hair reflections, and photographic tonal range. CAMERA: realistic smartphone front-camera/selfie perspective, close camera distance, approximately 24–28mm equivalent wide-angle perspective without excessive distortion, eye-level camera height, natural perspective, sharp facial detail, soft background separation, high dynamic range, realistic exposure. COMPOSITION: three equal-width horizontal portrait panels stacked vertically, thin white separators between panels, no extra borders or captions, consistent subject scale and environment continuity, carefully matching the reference framing and cropping. The final image should feel like an authentic candid social-media photo collage rather than an artificial composite. Preserve the user's identity consistently across all three panels while transferring only the reference subject's expressions, gaze directions, head poses, body/hand poses, clothing, environment, lighting, composition, and visual styling. Do not introduce additional people, objects, text, logos, watermarks, or unrelated elements. Maintain photorealism, natural anatomy, realistic hands and fingers, accurate facial proportions, detailed hair strands, subtle film-like warmth, soft cinematic contrast, and polished high-resolution photographic quality.

複製完整的提示並根據要求使用您自己的參考照片。

打開ChatGPT ↗

此提示提供的範例影像。此提示尚未經過我們網站的測試。

照相館
AI 產生的範例

為您自己的照片提供一點靈感。

1加入您的照片
A 人

保持臉部清晰可見

B 人

保持臉部清晰可見

JPG、PNG 或 WebP · 最大。每個 10 MB

僅當您選擇「產生」時才會發送您的照片。使用您有權使用的照片。

2讓它成為你的
更改詳情
編輯您的提示
開放雙子座

準備好…

登入您的帳戶

繼續使用您的 Google 帳戶。

您無需登入即可瀏覽和複製提示。

我們如何使用您的資訊: 隱私權政策