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

Ảnh chụp flash góc nhìn thứ nhất về người phụ nữ ngà ngà say trên phố đêm Nhật Bản

First-Person Flash Snapshot of a Tipsy Woman on a Late-Night Japanese Street

Ảnh chụp flash góc nhìn thứ nhất về người phụ nữ ngà ngà say trên phố đêm Nhật Bản — 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.

Ảnh chụp flash góc nhìn thứ nhất về người phụ nữ ngà ngà say trên phố đêm Nhật Bản — AI image creation
Ảnh chụp flash góc nhìn thứ nhất về người phụ nữ ngà ngà say trên phố đêm Nhật Bản

Prompt

Prompt tiếng Trung

使用上传的身份参考图作为唯一主可见20岁成年日本女性的面部身份依据，保留可辨认面部特征。创建9:16自然智能手机第一人称快照：深夜离开居酒屋后的日本城市街道，使用iPhone内置闪光灯；一手行走拍摄带轻微抖动，不是摆拍、纪念照、影棚照或商业照。闪光清晰呈现皮肤纹理、轻微不对称、细雀斑、自然瑕疵，以及饮酒后的脸颊和鼻尖泛红。 她穿适合深夜居酒屋归途的紧身圆领T恤和自然休闲装，波波头、白皙肤色、平均体型。她意识清醒，能自行走路，但酒后略摇晃、眼皮沉重、眼睛微湿、双颊鼻尖耳周泛红，带略困、轻松、柔和的闭口微笑。两人同向走路；摄影者从后方用整条手臂自然环住并支撑她的肩、上背和外侧腋下，部分分担不稳时的重量，绝不掐肩、勒脖或让她失去意识。摄影者脸和完整身体不可见，仅在边缘出现必要的衣物和支撑手臂。 相机在她面部上方、略靠前的第一人称位置，以约20-30度正面侧方的斜上视角拍摄。她躯干维持行进方向但轻靠摄影者，只有颈脸自然转向镜头；她仅抬眼看向更高镜头，头不大幅后仰。脸和上身约占画面70-80%，头到腰约80-90%画面高度；她脸从顶向中央出现，肩、上身和躯干向下，背景只留边缘。摄影者不要遮挡她或抢占主体。 背景是行人稀少的日式居酒屋/餐馆街：暗人行道、关门招牌、店面、自动售货机、护栏、电线杆、路灯、远处零星行人和出租车灯。闪光从斜上前方照亮她脸、发、肩与上身，眼中有小眼神光，额头鼻梁颧骨和唇有高光；主体与支撑臂明亮，数米外背景迅速暗1-3档但不完全黑。保留夜色、有限动态、轻微局部高光、适量运动感而闪光冻结面部；不用纯自然夜光、日景、过度HDR、柔棚灯、红眼、过曝、脸暗或背景均匀亮。避免第三方视角、面对面站位、远距离/全身/纯侧脸、女性很小、摄影者脸、断裂手臂或不自然肩颈、呕吐、失去意识、无法行走、额外人物、文字、Logo、水印和过度性化身体。

Prompt tiếng Anh

Use the uploaded identity reference as facial identity for the only main visible 20-year-old adult Japanese woman, preserving recognizable features. Create a natural 9:16 first-person smartphone snapshot on a Japanese city street after leaving an izakaya late at night, using iPhone built-in flash. It has slight one-handed walking shake and is neither posed commemorative, studio, nor commercial. Flash clearly shows real skin texture, mild asymmetry, faint freckles, natural imperfections, and drinking flush at cheeks/nose. She wears a fitted crew-neck T-shirt and natural casual clothing suitable for late-night izakaya return, with bob hair, fair skin and average build. She is conscious and independently walking but slightly unsteady from drinking, with heavy lids, slightly moist eyes, flushed cheeks/nose/ears and a sleepy relaxed gentle closed-mouth smile. Both walk same direction; photographer uses a full arm from behind to naturally support her shoulder, upper back and outer underarm, partly bearing weight when unsteady, never pinching shoulder, holding neck or making her unconscious. Photographer face/full body remain invisible; only necessary edge clothing and supporting arm appear. Place camera slightly ahead of and above her face in first person, at a 20-30-degree near-front side diagonal from above. Her torso faces walking direction but leans slightly to support; only neck/face turn naturally toward camera. She lifts only eyes toward higher camera, without strong head-back tilt. Her face/upper body occupy 70-80% of frame and head-to-waist about 80-90% height; face appears top-to-center, shoulders/upper body descend, and background stays at edges. Never let photographer obscure or outweigh her. Background is a sparse Japanese izakaya/restaurant street: dark sidewalk, closed signs/storefronts, vending machines, guardrails, poles, streetlights, scattered distant pedestrians and taxi lights. Flash from high-front lights face, hair, shoulders and upper body, with small eye catchlights and highlights on forehead, nose bridge, cheekbones and lips; subject/support arm are bright while background fades 1-3 stops within meters but never pure black. Preserve night, limited range, mild local highlights and walking energy while flash freezes face. No natural-light-only night, daytime, excessive HDR, soft studio flash, red-eye, overexposure, dark unreadable face or evenly lit background. Avoid third-person view, face-to-face position, distant/full-body/pure-side framing, tiny woman, photographer face, disconnected arm or unnatural shoulder/neck, vomiting, loss of consciousness, inability to walk, extra people, text, logo, watermark, or excessive sexualization.

Xem mô hình API

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