Galerie de prompts IA OpenZhenZhen · MiniMax H3 · video

Retrouvailles dans la ruelle sous la pluie

Reunion in the Rainy Alley

Retrouvailles dans la ruelle sous la pluie — découvrez cette création avec MiniMax H3 sur OpenZhenZhen, son prompt original et un exemple de génération à réutiliser.

Retrouvailles dans la ruelle sous la pluie — AI video creation
Retrouvailles dans la ruelle sous la pluie

Prompt

Prompt chinois

制作一段15秒、16:9的真人电影质感场景，地点是雨后的江南古镇。采用阴天柔和光线、低饱和的冷灰青色调与温暖木色，呈现真实的皮肤和织物纹理、轻微35毫米胶片颗粒及浅景深。场景中有两位成年女性；两人的外貌都以提供的身份参考图为准，服装则按以下描述保持。A女身穿淡蓝色传统长裙和白色外衫，站在湿润的石阶旁。B女身穿浅灰色传统长袍，佩戴精致的花形发饰和长耳坠。全程保持人物身份、位置、天气、光线和色调连贯。 摄影机固定，以缓慢、连续的光学变焦作为核心运镜，呈现自然的镜头压缩和焦点变化；不要使用数码变焦或移动摄影机。0至4秒，展现青石台阶、灰砖墙、木檐、雕花木门、竹帘和湿润巷道。A女沿台阶缓缓走到巷口，停下并望向廊檐下的B女。行人从镜头两侧近处经过，短暂形成柔和的前景遮挡，同时让A女仍可见。4至9秒，B女从右侧前景缓缓进入，在木柱旁停下，以木柱构成画框，让远处的A女留在画面中。继续缓慢光学变焦；A女逐渐虚化，B女的侧脸转为清晰，巷道纵深明显压缩，但两人的实际位置保持不变。9至15秒，B女转向A女。再以极缓慢的光学推近拍摄B女中近景，画面左缘保留一部分虚化的A女。短暂停顿后，B女轻声用中文说：“好久不见。”并露出浅浅笑意。让竹帘随微风轻动，屋檐偶尔滴水；以这段含蓄的重逢收尾。动作克制自然，曝光稳定，焦点转换平滑。避免现代物件、醒目的新增人物、人物重复、身份或服装漂移、肢体异常、摇摄或跟拍、突然的变焦或焦点切换、画面闪烁、过度雾化或磨皮，以及塑料质感的皮肤。

Prompt anglais

Create a 15-second, 16:9 live-action cinematic scene in a historic Jiangnan town just after rain. Use soft overcast light, muted cool gray-green tones balanced by warm wood, realistic skin and fabric textures, subtle 35mm film grain, and shallow depth of field. Include two adult women; use the supplied identity reference image for the appearance of each woman, while keeping their clothing as described here. Woman A wears a pale blue traditional dress with a white outer layer and stands by wet stone steps. Woman B wears a light gray traditional robe with delicate floral hair ornaments and long earrings. Keep their identities, positions, weather, lighting, and color consistent throughout. Use a fixed camera and one slow, continuous optical zoom as the central camera movement, with natural lens compression and focus shifts; do not use digital zoom or camera travel. From 0–4 seconds, establish the stone steps, gray brick walls, wooden eaves, carved door, bamboo blind, and wet alley. Woman A slowly descends to the lane entrance, stops, and looks toward Woman B beneath the eaves. Passersby cross close to the lens from either side, briefly creating soft foreground occlusions while Woman A remains visible. From 4–9 seconds, Woman B enters slowly from the right foreground and pauses beside a wooden pillar, framing the distant Woman A. Continue the gentle optical zoom; let Woman A soften out of focus as Woman B’s profile becomes sharp, with the alley’s depth visibly compressed while their actual positions remain unchanged. From 9–15 seconds, Woman B turns toward Woman A. Make a final very slow optical push toward Woman B in a medium close-up, retaining a softly blurred part of Woman A at the left edge. After a brief pause, Woman B says in Chinese, “好久不见。” (“Long time no see.”), with a faint smile. Let the bamboo blind stir in the breeze and a few eave drops fall; end on the quiet exchange. Keep motion restrained and natural, exposure stable, and focus transitions smooth. Avoid modern objects, extra prominent characters, duplicated people, identity or clothing drift, unstable anatomy, camera pans or tracking, abrupt focus or zoom changes, flicker, excessive haze or retouching, and plastic-looking skin.

Thèmes et styles

Voir le modèle API

Découvrir plus de prompts image et vidéo · Générer des images · Générer des vidéos · Créer sur le canevas IA