OpenZhenZhen AI 프롬프트 갤러리 · gpt-image-2.5 · image
새벽 3시의 과장된 광각 감자튀김 셀카
Exaggerated Wide-Angle French Fry Selfie at 3 AM
‘새벽 3시의 과장된 광각 감자튀김 셀카’은(는) gpt-image-2.5로 만든 작품입니다. OpenZhenZhen에서 원본 프롬프트와 생성 예시를 살펴보고 새로운 창작에 활용해 보세요.
프롬프트
중국어 프롬프트
创作一张竖幅9:16深夜手机自拍照：一位20岁的成年女性带着随性古怪的“怪女孩”风格，凌晨3点瘫坐在复古24小时快餐店的亮红色塑料卡座里。她直视手机闪光灯，表情厌倦而面无表情，轻咬一根薯条尖端，唇彩略显凌乱。凌乱头发塞在彩色针织帽下，大号有线耳机戴在帽子外；她穿宽松复古拉链连帽衫，上面别有各式珐琅徽章，并叠戴银戒指。采用高举手机的超广角自拍视角并呈现明显桶形畸变：前景一根薯条夸张地显得巨大，同时保持人物面部可辨。加入堆满汉堡包装纸、洒有番茄酱的红色塑料托盘，明亮无菌的荧光灯、模糊且无可读字样的霓虹窗牌，以及反射闪光的黑白方格油毡地板。采用刺眼的手机直闪观感与冷色荧光环境光；人物从杂乱背景中突出，卡座投下硬朗阴影，快餐红黄高饱和，并保留轻微桌面油渍、托盘眩光和阴影处数字噪点。整体呈现漫不经心、饥肠辘辘、混乱讽刺的Z世代照片堆风格，而非精致影棚肖像。可见成年女性外观绑定固定角色参考 ref-10a-v1。画面不出现其他人物、可读包装或招牌文字、水印，徽章上不出现标志。
영어 프롬프트
Create a vertical 9:16 late-night smartphone-selfie photo of one 20-year-old adult woman with an effortless, offbeat “weird-girl” style, slumped in a bright red plastic booth at a retro 24-hour fast-food restaurant at 3 AM. She looks straight into the phone flash with a bored, deadpan expression, lightly biting the tip of a French fry; her lip gloss is slightly messy. Her messy hair is tucked under a colorful knitted beanie, with large wired headphones worn over the beanie. She wears an oversized vintage zip-up hoodie covered with varied enamel pins and layered silver rings. Use a high-held ultra-wide selfie viewpoint with strong barrel distortion: make one foreground French fry look comically oversized while keeping her face recognizable. Include a red plastic tray crowded with burger wrappers and spilled ketchup, a bright sterile fluorescent ceiling, a blurred neon window sign with no legible wording, and a checkered linoleum floor reflecting the flash. Light with a harsh direct phone-flash look against cold fluorescent ambient light; let the subject pop from the clutter, with hard booth shadows, saturated fast-food reds and yellows, slight grease smears, tray glare, and digital noise in the shadows. Aim for a careless, hungry, chaotic Gen-Z photo-dump aesthetic, not a polished studio portrait. Anchor the visible adult woman's appearance to the fixed character reference ref-10a-v1. No extra people, readable packaging or signage, watermark, or logos on the pins.
더 많은 이미지·동영상 프롬프트 보기 · 이미지 생성 · 동영상 생성 · AI 캔버스에서 만들기