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

Phục hồi chân dung ánh sáng xiên giờ vàng

Golden Hour Side-Light Portrait Restoration

Phục hồi chân dung ánh sáng xiên giờ vàng — 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.

Phục hồi chân dung ánh sáng xiên giờ vàng — AI image creation
Phục hồi chân dung ánh sáng xiên giờ vàng

Prompt

Prompt tiếng Trung

以完整的已配置参考图 ref-10a-v1 作为原始图像,对这张人像进行修复。严格保留同一位成年女性的身份与自然年龄感、正面头肩姿势、略微分开的平静嘴唇表情、居中位置、原始裁切和横向构图、深棕色中分长直发、黑色细肩带上衣以及纯青绿色背景。不要改变姿势、表情、面部身份、发际线、衣物、身体、物体、裁切、几何关系或场景。为面部和头发增加温暖的黄金时段侧光,并让清晰而富有戏剧性的投影落在人物身后原有的纯青绿色背景表面上;保持背景简洁,不要添加建筑特征或新场景。使用暖色高光、深青绿色阴影、高对比度和自然的光线衰减,营造神秘的电影氛围。在保留可见毛孔和细腻自然纹理的同时,提升可感知的清晰度与真实皮肤细节;轻柔减少可见噪点,不要磨平面部,也不要产生光晕、过度锐化或塑料皮肤。保留参考图原有的图像尺寸和构图;只描述所需的视觉清理,不承诺超分辨率算法或输出分辨率。仅作写实修复:不要添加人物、道具、文字、标志或水印,不改变面部,也不重新构图。

Prompt tiếng Anh

Restore the actual supplied portrait using the complete configured reference image ref-10a-v1 as the source image. Strictly preserve the same adult woman's identity and natural age appearance, forward-facing head-and-shoulders pose, calm slightly parted-lip expression, centered placement, original crop and landscape framing, long straight dark-brown center-parted hair, black thin-strap top, and plain teal background. Do not change her pose, expression, facial identity, hairline, clothing, body, objects, crop, geometry or scene. Add warm golden-hour side lighting across the face and hair, with a defined dramatic cast shadow falling onto the existing plain teal background behind her; keep the source background plain and do not invent architectural features or a new setting. Use warm highlights, deep teal shadows, high contrast and natural light falloff for a mysterious cinematic mood. Improve perceived clarity and realistic skin-texture detail while retaining visible pores and fine natural texture; gently reduce visible noise without smoothing the face or introducing halos, crunchy sharpening or plastic skin. Preserve the reference's original image dimensions and composition; describe only the requested visual cleanup and do not promise an upscaling algorithm or output resolution. Photorealistic restoration only: no added people, props, text, logos, watermark, face changes or re-composition.

Xem mô hình API

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