VdoBloom
← Back

Surreal Immortal Mountain Environment Awakening

Short Filmby @Soran · source ↗August 12, 2026

Full prompt

Video Prompt | Optimized Structure Version
1) Overall Style
Cinematic realistic texture. Using "Reference-Driven Live World Design," blending poetic immortal-cultivation mystery, grand scale, restrained deadpan humor, natural volumetric air, and realistic textures of rock, water, vegetation, and silk. Core concept: The environment itself becomes the third dramatic character.
2) Pre-generation Environmental Planning
Perform silent environmental planning before generating. All reference images decide the environmental DNA: terrain, architecture, scale, material age, ecology, water system, weather, clouds, fog flow, lighting, color, reflections, and depth. Organize into a new, unified, physically reasonable landscape belonging to the same world. Principles: No direct copying of static backgrounds; lock terrain, architecture, water, and atmospheric anchors; maintain stable spatial relations across 3 shots; background must move naturally.
3) Character Setting
Character ID A | Senior Sister: Same 25-30 year old East Asian immortal from @Image1. Maintain: face, hairstyle, build, white embroidered silk Hanfu, jade accessories, white boots.
Character ID B | Junior Sister: Same 20-25 year old East Asian sister from @Image2. Maintain: face, hairstyle, build, turquoise Hanfu, dark shoes.
4) Shot Design
Shot 1 | 0–5s: Wide/Long shot. Two stand on a ridge. World moves naturally: water flows, mist drifts, trees sway, cloud shadows move, reflections change, distant buildings appear/disappear. Dialogue: Junior Sister whispers: "Senior Sister... is this mountain breathing?"
Shot 2 | 5–10s: Medium/Cowboy shot. White-clad immortal kneels, fingers on rock listening. Ground pulse spreads: grass ripples, water droplets tremble, birds fly, river reverses. She looks at the distant cliff.
Shot 3 | 10–15s: Close-up to focus reveal. Close-up of Junior Sister's boot; the "rock" rises slightly revealing scales. Focus shifts to distant cliff: A crevice opens revealing a golden giant eye. Trees and ruins are attached to this structure. The mountain is a creature. Dialogue: Junior Sister freezes: "Did we step on its tail?" Senior Sister: "Don't draw your sword yet." Pause: "Apologize first."
5) Global Constraints
Strict 15s total, 16:9 widescreen, 3 clear shots, stable relations, realistic physics for fog/wind/water, no subtitles.
6) Negative
blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, face morphing, background shift, glitching cuts, disappearing props

Original prompt

视频提示词|结构优化版
1)整体风格
电影级写实质感。
采用“参考图驱动的活世界设计”,融合诗意仙侠悬疑、宏大尺度、克制冷幽默、自然体积空气感,以及具有真实触感的岩石、水体、植被和真丝材质。
核心创意:环境本身成为第三个戏剧角色。
2)生成前的环境规划
正式生成前,先执行一轮无声环境规划。
由全部参考图共同决定环境 DNA,包括:
实际地形
建筑语言
空间尺度
材质年代
植被生态
水体系统
天气状态
云层高度
雾气流向
主光方向
整体色彩
反光表面
空气纵深
将以上信息重新组织为一个:全新、统一、物理合理、之前未出现过、属于同一世界的新景观。
环境构建原则:不可直接复制任何单张参考图作为静态背景;必须先确定并锁定地形锚点、建筑锚点、水体或植被锚点、大气锚点;三个镜头必须严格保持这些空间关系连续稳定;背景不能冻结,环境必须持续自然运行。
3)角色设定
角色 ID A|剑仙师姐:始终为 @图片1 中同一个 25–30 岁东亚剑仙女子。保持:同一张脸、同一发型、同一身材比例、白色刺绣真丝汉服、玉石饰品、白色布靴。
角色 ID B|小师妹:始终为 @图片2 中同一个 20–25 岁东亚小师妹。保持:同一张脸、同一发型、同一身材比例、青绿色汉服、深色布鞋。
4)镜头设计
Shot 1|0–5s:全景或远景。两人站在重新规划形成的同一条山脊上。世界持续自然运行:水体流动、山雾游走、树冠随风摇摆、云影慢移、反光改变、远处建筑隐现。对白:小师妹低声说道:“师姐……这山是不是在呼吸?”
Shot 2|5–10s:中景或牛仔景。白衣剑仙缓慢单膝俯身,手指贴在风化岩石表面倾听。地面脉冲沿环境传播:草丛倒伏波纹、叶尖水珠震颤、飞鸟腾起、河水逆流。她缓缓抬眼看向远处悬崖。
Shot 3|10–15s:特写或极近特写转焦点揭示。小师妹布靴特写,脚下“岩石”轻微起伏显露鳞片纹理。焦点转移至远处悬崖:悬崖裂隙睁开,显露金色巨眼。树木古迹始终依附在其结构上。山体即是生物。对白:小师妹全身僵住:“我们是不是踩它尾巴了?”剑仙平静回答:“先别拔剑。”停顿后补一句:“先道歉。”
5)全局硬约束
严格总时长 15 秒,16:9 横屏,三个镜头连续清晰,人物地理关系稳定,雾、风、水遵循物理逻辑,不生成字幕。
6)Negative
blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, face morphing, background shift, glitching cuts, disappearing props
Generate this with Seedance 2 on VdoBloom.
Opens the image-to-video creator with this prompt and Seedance 2 pre-selected.
✨ Generate Video