OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image

东京小巷成年女性与壁画之手合影

Adult Woman Posing with Mural Hand in a Tokyo Alley

A photorealistic 2:3 night-alley portrait of an adult woman beside a vivid anime-style mural whose hand reaches naturally onto her shoulder.

东京小巷成年女性与壁画之手合影 — AI image creation
东京小巷成年女性与壁画之手合影

提示词 / Prompt

中文提示词

创作一张 2:3 竖版全身中近景街头肖像，表现一位明确成年的 20 多岁女性站在东京深处的小巷。她拥有明亮象牙米色肌肤、自然细腻纹理、精致日系人偶风妆容、淡玫瑰腮红、香槟米色与粉棕眼影、纤长睫毛、柔和眉形、透明粉米色唇彩、清澈捕捉光和不过分甜腻的微笑。她穿夏日可爱日系休闲风：浅粉短袖开衫搭白色蕾丝边罗纹吊带背心，白至象牙色小花褶边迷你裙、白色绑带凉鞋和米色链条包。身体轻倚旧墙，一侧膝盖微弯，右手握包带，左手自然垂在裙侧。墙上是一幅高分辨率平面动画风涂鸦壁画：橙色头发的航海士风角色俯身，一只壁画手自然轻放在现实女性肩膀上，另一只手斜握长棍。接触点必须一眼可读且关系自然，壁画保持平面并紧贴墙面，不得立体化或跳出墙外。加入龟裂漆面、剥落墙壁、简朴日文招牌、灯笼、贴纸、左上方暖街灯、右下方柔和阴影以及橙色/青色/米白/淡粉色调。聚焦女性脸部和肩膀，同时以中等景深保持壁画线条清晰。避免幼态、不自然视线、手指错误、接触错位、壁画手臂弯曲异常、模糊低分辨率线稿、多余角色、文字、标志、过曝或水印。两种角色均使用配置的人物参考。

English prompt

Create a photorealistic 2:3 vertical full-body medium-close street portrait of a clearly adult woman in her 20s in a deep Tokyo alley. Give her bright ivory-beige skin with natural texture, refined Japanese-doll-inspired makeup, pale rose blush, champagne and pink-brown eyeshadow, long lashes, soft brows, transparent pink-beige gloss, clear catchlights, and a gentle not-too-sweet smile. Dress her in summer Japanese casual fashion: light pink short-sleeved cardigan over a white lace-edged ribbed camisole, white-to-ivory floral-ruffle mini skirt, white lace-up sandals, and small beige chain shoulder bag. Let her lean lightly against the old wall with one knee bent; her right hand holds the bag strap and left hand rests beside the skirt. On the wall behind her, make a large high-resolution flat anime-style graffiti mural of an orange-haired navigator character leaning down, one wall-bound illustrated hand gently touching her shoulder and the other holding a long staff diagonally. Keep the contact point instantly readable and physically coherent without making the mural 3D or allowing it to leave the wall. Include cracked paint, peeling wall, simple Japanese signs, lanterns, stickers, warm upper-left streetlight, soft lower-right shadows, and orange/cyan/off-white/pale-pink palette. Focus on face and shoulders with medium depth of field while keeping mural lines clear. Avoid childish appearance, unnatural gaze, bad fingers, misaligned contact, unnatural mural arm, blurry low-resolution art, extra characters, text, logos, overexposure, or watermarks. Use the configured identity reference for both roles.

查看 API 模型

探索更多图片和视频提示词 · 生成图片 · 生成视频 · 在 AI 画布中创作