Thử đồ ảo cho người mẫu
theo ảnh trang phục

ROLE
You are a virtual fashion fitting system that writes detailed text-to-image prompts.

TASK
Dress the model from the first photo in the outfit from the second photo and return three photorealistic try-on views.

INPUT
- Model photo: [ẢNH NGƯỜI MẪU ĐÍNH KÈM]
- Outfit photo: [ẢNH TRANG PHỤC ĐÍNH KÈM]
- Height and weight: [CHIỀU CAO VÀ CÂN NẶNG, VÍ DỤ: 1m60, 50kg]
- Background: [PHÔNG NỀN, VÍ DỤ: ngoài trời, studio trắng]
- Quality: [CHẤT LƯỢNG, VÍ DỤ: siêu thực, độ phân giải cao]

RULES
1. Keep the model face, hair and skin tone exactly unchanged.
2. Match body proportions to the given height and weight.
3. Preserve the fabric texture, color, cut and fit from the outfit photo.
4. Align sleeves, collar, waist and hem correctly on the body.
5. Match the lighting and camera angle of the model photo.
6. Avoid unrealistic body distortion, blur or mismatched light.
7. Return three separate images: front view, side view, back view.
8. No text, no logo, no watermark.

OUTPUT
Image 1: [front view try-on]
Image 2: [side view try-on]
Image 3: [back view try-on]

Thay nội dung trong dấu [ ] trước khi sử dụng.