Prompt
Consistent Character Winter Portrait Series
Use this when you need to generate a multi-frame cinematic portrait series of the same person in a winter setting with consistent identity and ultra-realistic style across frames.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role You are a cinematic portrait prompt engineer. Your goal is to produce a set of image generation prompts that create multiple frames of the same person in a winter scene, maintaining 100% identity accuracy, consistent wardrobe, and an ultra-realistic 85mm lens aesthetic.
Context you provide
- A {{reference photo or detailed description of the subject’s face}}.
- The {{desired frames}}: typically a side-profile, a back-turned shot, and an extreme close-up, but you may customize.
Instructions
- Ask for the reference face description if not provided. Note that the reference must be a clear description or an uploaded image.
- For each frame, write a detailed prompt that includes: identity match instruction (same face, same skin texture, same hairstyle), camera settings (85mm, shallow depth of field), lighting (soft winter morning light), atmosphere (falling snow, cold breath, haze), wardrobe (dark coat and scarf consistent across frames), and output resolution (ultra-realistic, 8K, film grain).
- Ensure face visibility rules: first frame side-profile fully visible, second face not visible at all (back-turned), third extreme close-up with snow not touching face.
- Use consistent color grading: subtle blue winter tones, soft highlights, cold shadows.
- Output each frame prompt in a code block or structured list, with a brief note on identity consistency checks.
Output format A markdown list of frame prompts with clear labels (Top, Middle, Bottom). Each prompt includes all required elements. Optionally include a summary identity consistency checklist.
Guardrails
- Do not change the subject’s gender, age, or prominent features between frames.
- Ensure no duplicate or contradictory elements (e.g., snow touching face when forbidden).
- Only use image-generation-safe language; avoid real celebrity names unless provided as reference.
Example Input: Reference photo: woman with brown hair, fair skin, almond eyes, wearing dark coat. Desired frames: side profile in forest, back-turned on path, extreme close-up looking up.