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

溪岸巨龟:古风人物与巨物相伴静帧

Giant Turtle by the Stream: Ancient-Style Figure with a Colossal Companion

A bilingual 16:9 cinematic still prompt with explicit Jiangnan stream-bank, giant-tortoise, gentle-touch interaction, and overcast-after-rain defaults. One clearly adult Hanfu woman is bound to the configured ref-10a-v1 identity reference at generation time; no reference transmission or image genera

溪岸巨龟:古风人物与巨物相伴静帧 — AI image creation
溪岸巨龟:古风人物与巨物相伴静帧

提示词 / Prompt

中文提示词

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。创作一张横向 16:9 的完整写实电影静帧,呈现中式巨物美学与东方诗意超现实气质,氛围宁静、舒展。场景:有低矮石岸与稀疏青苔的安静江南溪岸。天气与时段:小雨刚停、薄雾未散的阴天清晨。只出现一种主要动物:一只解剖结构真实的巨型乌龟,体长约为成年女子身高的十倍。画面中有一位明确成年的东方女子,身着符合历史感的汉服,人体比例真实,乌发简洁挽起、发饰克制,神情平静专注;不增加其他可见人物。她只进行这一项互动:女子将一只手掌轻放在可触及的龟壳下侧,乌龟随即温和地朝她转头。以中景清楚呈现可实际触及的接触位置。动物是画面情绪中心并主导构图;以动物体长作为尺度参照,约为女子身高的 8–15 倍,同时保留该物种真实的骨骼、关节、身体比例、重量感和自然头颈连接。动物体量巨大但不怪兽化、不拟人化。完整保留其头部、接触点及足以读懂互动的身体姿态,其余躯干可延伸出画框。人物与动物处于相近景深,直接呈现清楚的体量对比。摄影机位于人物活动高度,以适度广角的中远景平视或轻微仰视拍摄。前景放少量雨后湿石与稀疏青苔,中景清楚呈现手掌与龟壳的接触,远景以柔和溪岸和空气透视延伸纵深。人物双脚稳稳落地;触摸处有可信的接触阴影与轻微受力痕迹。龟壳呈现真实弧面盾片、细微磨损、含蓄潮湿反光和自然的深橄榄褐色变化,解剖结构清楚可辨为龟类。女子穿月白与浅青的棉麻丝质汉服,布料有重量和自然衣褶,以少量朱红点缀;衣料顺着姿态和微风轻动。雨后阴天的清晨侧光柔和地勾勒龟壳与衣裙,亮部保留纹理、暗部保留层次;反射克制,水线自然,远处只留薄雾。只呈现一个安静、可信的日常瞬间,画面开阔留白,整体达到写实电影静帧质感。不要在龟壳上添加建筑、奇幻盔甲或法术光环,不要攻击或恐怖氛围、动物拟人表情、不可触及的接触、失重悬浮、肢体融合或多余肢体、错误关节、毛绒玩具或塑料表面、现代物件、文字、字幕、标志或水印。

English prompt

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. Create one complete, realistic cinematic still in horizontal 16:9 format, with a serene Chinese giant-beast aesthetic and Oriental poetic surrealism. Scene: a quiet Jiangnan stream bank with low stones and sparse moss. Weather and time: a misty overcast morning just after light rain. Show exactly one main animal: one anatomically realistic giant tortoise, with body length about ten times an adult woman's height. Include one clearly adult Oriental woman in historically grounded Hanfu, with realistic adult proportions, black hair pinned simply, restrained adornment, and calm, attentive expression; no additional visible characters. She performs exactly this interaction: the woman rests one palm against the reachable lower side of its shell as the tortoise gently turns its head toward her. Make the contact physically reachable and clearly visible at medium distance. The animal is the emotional center and dominates the frame; use its body length as the scale reference, about 8–15 times the woman's height, while retaining the selected species' real skeletal structure, joints, proportions, weight, and natural head-to-neck connection. Keep the animal massive but not monstrous or anthropomorphic. Preserve its complete head, contact point, and enough body posture to read the interaction; its remaining torso may extend beyond the frame. Place woman and animal at comparable spatial depth so their scale contrast is immediately clear. Use a moderate wide-angle medium-to-long view from the woman's activity height, eye level or slightly upward. Build a small foreground of damp stones and sparse moss, a clear middle ground containing the hand-to-shell contact, and a soft streambank background with atmospheric depth. The woman's feet remain firmly grounded; show plausible contact shadow and slight compression at the touch point. Render the tortoise shell with realistic curved scutes, fine wear, subtle damp reflections, and natural dark olive-brown variation; keep anatomy recognizably tortoise-like. Dress the woman in moon-white and pale-cyan cotton-silk Hanfu with weighted folds and a small vermilion accent; fabric follows her posture and local breeze. Soft overcast morning side light after rain outlines shell and garment, retaining highlight texture and shadow detail; use restrained reflections, natural water lines, and only a trace of distant mist. Keep one quiet daily-life moment, tender and credible, with open negative space and a photorealistic cinematic finish. Do not add buildings on the shell, fantasy armor, magic halos, attack or horror, human-like animal expressions, unreachable contact, floating weight, merged limbs, extra limbs, wrong joints, plush or plastic surfaces, modern objects, text, subtitles, logos, or watermarks.

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

  • 花市试戴宫衣发簪的镜前一刻

    人物身穿半臂宫衣,在花市挑选发簪并试戴:微微低头整理发簪,再抬眼看向镜中,唇角轻扬。清晰肩线、自然贴合身形的衣料、显露的胸前肌肤、纤细腰身与明确腰胯线条是服装造型重点。

  • 黑色背景下的戏剧性黑白人像

    以固定人物参考保持单一成年女性的可识别身份,创作居中的纵向黑白美术肖像。她穿黑色高领上衣,置身近乎纯黑的极简影棚背景;单侧柔光形成强烈明暗对照,保留自然皮肤质感、细腻灰阶和克制胶片颗粒。

  • 透明亚克力手绘餐饮门店标牌

    正面平视展示的透明亚克力餐饮标牌:虚构牛肉面馆品牌、手绘店名和标语、牛肉面与小图标以彩色线稿绘于面板上;温暖店内环境透过透明区域可见。

  • 暗黑奇幻新可爱风成年角色吉祥物设定

    将指定成年暗黑奇幻角色压缩为轮廓圆润、信息精炼的新可爱风吉祥物:保留发型、角翼尾耳、眼睛、主色和标志性服装配件等身份锚点,以干净白或浅灰背景、大留白和精炼商业插画线条呈现危险感与可爱感并存的角色设定海报。

  • 露营车内夜间放松的宽景人像

    夜间露营车内的 3:4 竖幅宽景:一位成年人物放松地坐在侧椅上,视线不直视镜头;画面不出现外套或可见标志。A wide 3:4 nighttime view inside a camper van, with an adult subject relaxing on a side chair and looking away from the camera.

  • Y2K风格韩系女性数码相机镜前自拍

    一位明确成年的韩国女性在色彩丰富的千禧年代初期卧室中,用小型银色数码相机拍摄自然镜前自拍;保留指定Y2K穿搭、直闪、轻微过曝与真实数码快照质感。