Create a Candid Couple Selfie Inside an Indian Train

[ROLE]
Expert AI Image Generation Prompt Engineer and Photorealism Specialist

[CONTEXT]
You are generating a detailed, highly accurate text-to-image prompt designed for advanced AI image generators (such as Midjourney, DALL-E 3, or Stable Diffusion) to produce an authentic, candid photograph of a young South Asian couple inside an Indian train.

 [OBJECTIVE]
Rewrite and structure the user's raw prompt into a professional, highly optimized, production-ready image generation prompt that ensures strict adherence to facial identity, precise lighting, authentic cultural details, and a candid smartphone aesthetic.

[REQUIREMENTS]
- Subject: A young South Asian couple (woman on the left, man aged 18-20 on the right) inside an Indian commuter train compartment.
- Woman's Appearance & Attire: Long dark hair swept over one shoulder, distinct dark bindi, red lipstick, white block-print top with a blue paisley pattern and a keyhole neckline with a tie, white wired earphones in her ears. Posing by playfully resting her head lightly against the man's shoulder with a soft, natural smile looking toward the camera.
- Man's Appearance & Attire: Aged 18-20, facial stubble, open-collar black button-down shirt, thin silver chain. Hairstyle must match the provided hairstyle reference image exactly. Posing relaxed, slightly turned toward her, gently holding the earphone wire near her shoulder with one hand, looking directly at the camera with a calm expression.
- Facial Identity Lock: Strictly maintain the man's original face, identity, facial features, facial structure, and overall appearance from the reference image.
- Lighting & Atmosphere: Strong, warm, golden-hour sunlight streaming through the window on the left, casting intricate patterns of light and shadow across the woman's face and chest. Natural film grain, realistic skin textures, authentic proportions.
- Environment: Utilitarian Indian train interior featuring metal luggage racks, circular ceiling fans, ventilation grilles, and small skylight windows on the ceiling.
- Camera & Composition: Eye-level smartphone selfie perspective, candid aesthetic, shallow depth of field keeping the couple sharp while the background fades slightly, seamless integration.

[CONSTRAINTS]
- Do NOT alter the man's facial features or identity from the reference image under any circumstances.
- Do NOT use overly glossy or plastic skin smoothing; prioritize natural skin textures and realistic imperfections.
- Do NOT make the lighting flat; the golden-hour side-lighting and shadow patterns are mandatory.

[OUTPUT FORMAT]
Provide a single, beautifully structured, comma-separated and descriptive image generation prompt block optimized for maximum AI rendering accuracy, followed by the aspect ratio parameter.

[QUALITY CHECKLIST]
- Are all subject details (clothing, accessories, positioning) preserved?
- Is the golden-hour lighting explicitly emphasized?
- Is the face-lock constraint for the man clearly established?
- Is the aspect ratio (3:4) correctly specified?
It should not look like AI-generated, it should looks natural — for a photographic image, like it was taken with an iPhone 12 camera. Generate image now