Convert the following T2I image generation prompt into a final prompt for an image generation model.

CRITICAL RULES:
1. The image generation model does NOT know any character names, location names, or proper nouns
2. Replace ALL character names with generic visual descriptions (e.g., "a young Korean man", "an elderly woman")
3. Replace ALL location names with their visual descriptions only
4. Replace ALL object names with generic descriptions
5. Translate to English — EXCEPT: proper nouns (place names, business names, etc.) should preferably stay in original language. Also keep any expression in original language where visually natural or where English translation would be awkward
6. Remove ALL marker brackets: [[]], [:]
7. Do NOT include camera angles or shot types (no "close-up", "wide shot", "medium shot")
8. Keep "Photorealistic cinematic still." at the beginning

OUTPUT FORMAT (this exact structure):

First, reference image descriptions (use the SAME visual descriptions you used in the scene):
{ref_roles_text}

Then the scene description.

Then at the end, the generation instructions:
Generate one image:
{ref_instructions}
- CRITICAL: Do NOT alter the face, age, hair, or identity from reference images. The face in the reference defines the character's exact appearance regardless of any outfit or age description in the scene.

World/style context to incorporate as visual atmosphere (NOT as named locations):
{style_context}

Original T2I prompt to convert:
{t2i_prompt}

Output ONLY the final English prompt. No explanations.