Thư viện prompt AI OpenZhenZhen · gpt-image-2.5 · image

Hai người bạn trò chuyện bên cửa sổ nhà hàng

Two friends chatting by a restaurant window

Hai người bạn trò chuyện bên cửa sổ nhà hàng — khám phá tác phẩm được tạo bằng gpt-image-2.5 trên OpenZhenZhen, kèm prompt gốc và ví dụ tạo nội dung để tham khảo.

Hai người bạn trò chuyện bên cửa sổ nhà hàng — AI image creation
Hai người bạn trò chuyện bên cửa sổ nhà hàng

Prompt

Prompt tiếng Trung

以用户上传的参考图作为主要可见人物的面部身份依据。在不与身份参考冲突的情况下,保留来源的场景、造型、服装、姿势、构图、光线和视觉风格。创作9:16 竖幅自然街拍,从麦当劳临街玻璃窗外向内看到窗边柜台区域。两位成年女性朋友端正放松地坐着聊天;主位女性面部遵循 ref-10a-v1,另一位同伴是面容明显不同的虚构成年女性。两人穿端庄日常上衣与过膝半裙,姿态舒适,双脚自然落在地面。手边有两杯外带冰咖啡,手部自然可见,神情投入而不摆拍。镜头在公共人行道上保持礼貌距离,玻璃只有轻微反射,不遮挡面部。室内呈现少量可辨餐桌和温暖灯光,不出现清晰可读的商标,也不加入无关顾客。自然傍晚环境光,真实手机镜头透视,9:16构图平衡。不要窥视式角度、性化姿势、内衣外露、刻意强调腿部或可识别的路人。

Prompt tiếng Anh

Use the uploaded reference image as the facial identity reference for the main visible person. Keep the source scene, styling, clothing, pose, composition, lighting and visual style unless they conflict with the identity reference. Make a 9:16 vertical candid street photograph looking through the clear front window of a McDonald's into its window-side counter area. Two adult women friends sit upright across the counter, chatting naturally; the main woman follows ref-10a-v1, while her companion has a clearly distinct fictional face. Both wear modest everyday tops and below-knee skirts, with relaxed posture and feet resting naturally on the floor. Their hands are visible near two takeaway cups of iced coffee; expressions are engaged and unposed. Frame them from the public sidewalk at a respectful distance, with subtle window reflections that do not obscure their faces. Show a few recognizable restaurant tables and warm interior lights, without legible logos or unrelated customers. Natural evening ambient light, realistic phone-camera perspective, balanced 9:16 composition. No voyeuristic angle, sexualized pose, exposed underwear, leg-spread emphasis, or identifiable bystanders.

Xem mô hình API

Xem thêm prompt hình ảnh và video · Tạo hình ảnh · Tạo video · Tạo trên canvas AI