VdoBloom
← Back

Cinematic Wuxia comedy swordmaster scene

Short Filmby @Soran · source ↗August 13, 2026

Full prompt

[Generation Mode and Total Requirements]

Must use Seedance 2.0 Fast, strictly generate a full 15-second video, 16:9 widescreen, three continuous clear shots, with native synchronized Mandarin dialogue.

Master-level cinematic realistic texture, Arri Alexa film camera look, with a restrained and poetic Xianxia deadpan comedy temperament. The image is sharp and clear, volumetric morning light is naturally visible, acting is restrained, character movement is realistic but cinematic, and silk fabric, hair, paper, water flow, mountain mist, and natural light have believable physical feedback.

Adopting the 'Reference Image-Driven Living World System': All reference images together determine the location and visual world, including topographic logic, architectural language, spatial scale, material age, vegetation ecology, water flow direction, weather state, cloud movement, main light direction, reflective surfaces, comprehensive colors, and atmospheric depth. Before official generation, complete a round of environmental planning silently, reorganizing a new, unified, physically reasonable space from these common environmental DNAs.

The background must be vivid, unique, and continue to operate naturally, but only responsible for environmental presence, not actively triggering punchlines, changing character goals, or stealing the center of the plot.

[Characters and Core Props]

Character ID A | Swordmaster Senior Sister
Always the same 25-30 year old East Asian female from @Image 1, strictly maintaining the same face, oval face shape, fair natural skin tone, dark almond eyes, black long hair half-tied, white jade hairpin, tall and slender body proportions, white embroidered silk Hanfu, semi-transparent layered wide sleeves, silver waist seal, jade pendant, and white cloth boots.
Character ID B | Junior Sister
Always the same 20-25 year old East Asian female from @Image 2, strictly maintaining the same face, rounded and lively face shape, black hair braided, petite body proportions, green linen Hanfu, dark waist belt, wooden hairpin, and black cloth shoes.

Core Prop
Only one and the same sealed bamboo scroll throughout. The position, opening state, and holding relationship of the bamboo scroll in the three shots must be continuous and stable.

[0–5s | Low-angle panorama, slow track]

The mountain gate corridor stone platform on a morning after rain. Below is a sea of clouds and a valley, carved stone railings, wet slate, narrow canals, ancient pines, light-colored flowering branches, distant tiled palaces, slowly wandering mountain mist, and continuously changing wetland reflections together establish a real, breathing environment.
The same white-clad swordmaster senior sister and the same green-clad junior sister are located on both sides of the stone table.

The white-clad senior sister gently puts the same sealed bamboo scroll on the stone table between them, saying calmly:

'Three years ago, you left something behind. Today it is returned.'

The environment maintains restrained motion: the canal flows continuously, the mist slowly passes through the corridor, branches and hems respond to the same morning breeze, and the reflection on the stone surface after the rain changes naturally with the camera track.

[5–10s | Cowboy shot within a shot]

The same green-clad junior sister slowly reaches out to take the same bamboo scroll, while her other hand naturally rests by her sword.

The same white-clad senior sister maintains a completely calm expression and adds a sentence:

'I have already graded it for you.'

Nearby leaves naturally drift past the screen, forming realistic foreground parallax; the canal continues to flow, and mountain mist slowly wanders between the corridors.

The background always maintains only a sense of world presence and does not actively create new plot events.

The junior sister starts to unfold the bamboo scroll after taking it, maintaining continuous action for the next shot.

[10–15s | Close-up to extreme close-up]

The same bamboo scroll is fully unfolded.

Inside is not a secret manual, but an old love poem written by the same green-clad junior sister three years ago; the paper is densely written with neat, elegant red annotations.

The same green-clad junior sister freezes instantly and asks in a low voice:

'You read it all?'

The same white-clad senior sister answers calmly:

'I also changed three rhyme schemes; the third line is the best.'

An extreme close-up captures the junior sister's ears turning noticeably red.

The senior sister remains dignified, calm, and without any hint of mockery.

At this moment, a drop of rain from the eaves naturally falls into the same canal, making a slight and clear water sound, as the final deadpan beat.

Strictly ends at 15.0 seconds.

[Execution Hard Constraints]

Must use Seedance 2.0 Fast
Must generate full 15 seconds, no shortening to 10 seconds, no early ending
16:9 widescreen
Strictly three shots
Native synchronized Mandarin dialogue
Accurate Mandarin lip-syncing
Character identity, face, hairstyle, body proportions, clothing stable throughout
The same bamboo scroll is continuous and stable, no duplication, no disappearing, no replacement
Foreground, midground, and background have realistic parallax
Environment continues to live, but must not steal the scene
Silk, paper, water flow, mountain mist, hair, flower branches, raindrops, and light are physically real
Only two visible characters appear at all times
No screen subtitles generated
No extra characters appear

[Negative]
blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, subtitles, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, face morphing, hairstyle change, background shift, architecture mutation, lighting jump, glitching cuts, disappearing props, duplicated bamboo scroll, changing scroll design, extra characters, background crowd, environment stealing focus, random environmental gag, exaggerated comedy acting, slapstick reaction, overacting, modern elements, frozen background, static fog, static water, no parallax, abrupt camera jump, dialogue desync, lip-sync error

Original prompt

【生成模式与总要求】

必须使用 Seedance 2.0 Fast 生成,严格生成完整 15 秒视频,16:9 横屏,三个连续清晰镜头,原生同步普通话对白。

大师级电影写实质感,Arri Alexa 电影机观感,带克制而诗意的仙侠冷面喜剧气质。画面锐利清晰,体积晨光自然可见,表演克制,人物运动真实但具有电影感,真丝衣料、头发、纸张、水流、山雾与自然光具有可信物理反馈。

采用“参考图驱动的活世界系统”:所有参考图共同决定地点与视觉世界,包括地形逻辑、建筑语言、空间尺度、材质年代、植被生态、水流走向、天气状态、云雾运动、主光方向、反光表面、综合色彩和空气纵深。正式生成前先无声完成一轮环境规划,从这些共同环境 DNA 中重新组织出一个全新、统一、物理合理的空间。

背景必须鲜活、独特并持续自然运行,但只负责环境存在感,不主动触发包袱,不改变人物目标,不抢夺剧情中心。

【角色与核心道具】

角色 ID A|剑仙师姐
始终是 @图片1 中同一个 25–30 岁东亚女性,严格保持同一张脸、椭圆脸型、白皙自然肤色、深色杏眼、黑色长发半挽、白玉簪、高挑纤细身材比例、白色刺绣真丝汉服、半透明分层宽袖、银色腰封、玉佩和白色布靴。
角色 ID B|小师妹
始终是 @图片2 中同一个 20–25 岁东亚女性,严格保持同一张脸、圆润灵动脸型、黑发编辫、娇小身材比例、青绿色亚麻汉服、深色腰带、木簪和黑色布鞋。

核心道具
全程只有同一卷密封竹简。竹简在三个镜头中的位置、开合状态和人物持有关系必须连续稳定。
【0–5s|低角度全景,缓慢推轨】

雨后清晨的山门回廊石台。下方是云海山谷,雕花石栏、湿润石板、狭窄水渠、古松、浅色开花枝条、远处瓦顶殿宇、缓慢游走的山雾与持续变化的湿地倒影共同建立一个真实存在、持续呼吸的环境。
同一个白衣剑仙师姐与同一个青衣小师妹位于石案两侧。

白衣剑仙师姐把同一卷密封竹简轻轻放到两人之间的石案上,平静说道:

“三年前,你落下一物,今日归还。”

环境保持克制运动:水渠持续流动,薄雾缓慢穿过回廊,树枝 and 衣摆响应同一晨风,雨后石面反射随摄影机推轨自然变化。

【5–10s|牛仔景中景】

同一个青衣小师妹缓慢伸手去取同一卷竹简,另一只手仍自然停在剑旁。

同一个白衣剑仙师姐保持完全平静的神情,补上一句:

“我已替你批过。”

近处叶片自然掠过画面,形成真实前景视差;水渠持续流动,山雾在回廊之间缓慢游走。

背景始终只维持世界存在感,不主动制造新的剧情事件。

小师妹取到竹简后开始展开,为下一镜头保持连续动作。

【10–15s|特写转极近特写】

同一卷竹简完整展开。

里面并不是秘籍,而是同一个青衣小师妹三年前写下的一首旧情诗;纸面上密密写满工整、优雅的红色批注。

同一个青衣小师妹瞬间僵住,低声问:

“你全看了?”

同一个白衣剑仙师姐平静回答:

“还改了三处韵脚,第三句最好。”

极近特写捕捉小师妹耳尖明显泛红。

剑仙师姐依旧保持端庄、平静、毫无取笑意味的神情。

此时檐角一滴雨水自然落入同一条水渠,发出一声轻微而清楚的水响,作为最后的冷面节拍。

严格在 15.0 秒结束。

【执行硬约束】

必须使用 Seedance 2.0 Fast

必须生成完整 15 秒,不缩成 10 秒,不提前结束

16:9 横屏

严格三个镜头
原生同步普通话对白
普通话口型准确
人物身份、面部、发型、身体比例、服装全程稳定
同一卷竹简连续稳定,不复制、不消失、不更换
前中后景具有真实视差
环境持续活着,但绝不能抢戏
真丝、纸张、水流、山雾、头发、花枝、雨滴和光线物理真实
始终只出现两名可见角色
不生成画面字幕
不出现额外人物
【Negative】
blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, subtitles, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, face morphing, hairstyle change, background shift, architecture mutation, lighting jump, glitching cuts, disappearing props, duplicated bamboo scroll, changing scroll design, extra characters, background crowd, environment stealing focus, random environmental gag, exaggerated comedy acting, slapstick reaction, overacting, modern elements, frozen background, static fog, static water, no parallax, abrupt camera jump, dialogue desync, lip-sync error
Generate this with Seedance 2 on VdoBloom.
Opens the image-to-video creator with this prompt and Seedance 2 pre-selected.
✨ Generate Video