VdoBloom
← Back

Xianxia Zither Crane Feeding Comedy

Cinematic Scene ShowcaseShort Filmby @Soran · source ↗July 28, 2026

Full prompt

Poetic Wuxia × Deadpan Comedy | The Cranes Recognize Rice, Not the Zither
[Style]
Cinematic photorealism, live-action realism. A fusion of restrained poetic wuxia aesthetics, deadpan observational comedy, elegant classical Chinese composition, soft evening backlighting, low-saturation jade green and ivory white tones, fine film grain, and controlled group coordination of spiritual cranes. The core of the reversal: Animal instincts easily defeat immortal majesty. The first half creates a high-class, solemn, and ethereal spectacle of "zither music attracting cranes," while the second half uses a tiny daily accident for a deadpan teardown. The entire film is set in a pure ancient Chinese world, containing no modern elements.
[Generation Focus]
Optimized for Seedance 2.0 capabilities: Reference image identity locking, multi-shot continuous narrative, dual character stability, group animal coordination, and native audio-visual synchronization.
[Duration] Strictly 10 seconds.
[Aspect Ratio] 16:9 widescreen, 24fps.
[Shot Structure] Three clean and clear shots.
I. Character Locking
Character A | Sword Immortal Senior Sister: Locked via @Image 1 (face, hairstyle, body proportions, clothing, shoes, accessories). Setting: 25–30 year old East Asian female, oval face, fair natural skin tone, dark almond eyes, black long hair half-tied with a white jade hairpin, tall and slender. Clothing: White embroidered silk Hanfu, layered translucent wide sleeves, silver waist seal, jade pendant, white cloth boots. Temperament: Cold, calm, dignified, confident, striving to maintain immortal majesty.
Character B | Junior Sister: Locked via @Image 2. Setting: 20–25 year old East Asian female, round and lively face, black hair braided, small build. Clothing: Cyan linen Hanfu, dark cloth belt, wooden hairpin, black cloth shoes. Prop: Holding a small bamboo basket of grain. Temperament: Natural, sincere, admires Senior Sister, deadpan calm.
II. Animal Locking
Crane Group: Strictly 3 white cranes throughout. Must maintain white feathers, slender physique, spiritual temperament, and natural movement. First half: elegant and neat. Second half: instinctive, eager, and noisy due to grain.
III. Scene Setting
A stone platform on a cliff edge above a sea of clouds. Includes: ancient pine, stone railing, wooden pavilion, bronze incense burner, prayer banners, flagstone ground, sea of clouds, mist, and an ancient zither. Fixed spatial relationship: Senior Sister at the zither, Junior Sister by the pavilion, cranes circling then landing behind Senior Sister, grain spilling between the zither and Junior Sister.
IV. Shot Design
Shot 1 (0–3s): Low-angle horizontal tracking full shot. Establishes the stable relationship between characters, zither, cranes, and pavilion. Senior Sister plays zither while 3 cranes circle above. Sound: Zither, wind, distant crane cries, silk friction.
Shot 2 (3–6s): Medium side shot. Senior Sister finishes a phrase and says: "When the zither plays, a hundred birds pay homage." The 3 cranes land gracefully behind her. Junior Sister shows admiration. Sound: Zither tail, wings, Mandarin dialogue with accurate lip sync.
Shot 3 (6–10s): Close-up reveal to extreme close-up. Junior Sister's basket tilts, grain spills. The 3 cranes instantly abandon formation, rushing past Senior Sister to peck at the grain. One crane steps on the zither, creating a discordant note. Junior Sister says expressionless: "Senior Sister, they recognize the rice." Extreme close-up of Senior Sister frozen, eyebrow twitching, a white feather landing on her hairpin. Sound: Grain spilling, crane footsteps, pecking, discordant note, Mandarin dialogue.
V. Action Sequence (Last 4s): 1. Basket tilts -> 2. Grain spills -> 3. Cranes attracted -> 4. Cranes rush -> 5. Crane steps on zither -> 6. Junior Sister speaks -> 7. Cut to extreme close-up -> 8. Senior Sister freezes/eyebrow twitch -> 9. Feather falls.
VI. Performance Requirements: Senior Sister must be truly elegant for 6s, then show a silent breakdown. Junior Sister stays calm and sincere. Cranes transition from elegant to instinctive.
VII. Physics/Continuity: Natural movement for wings, fabric, and grain. Strictly 3 cranes, no extra people, stable character features and scene positions.
VIII. Sound: Native synced Mandarin, cinematic environment layers, no subtitles, no modern sounds.
IX. Technical: 10s, 3 shots, 16:9, 24fps, Seedance 2.0 locking.
X. Negative Prompts: blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, subtitles, captions, logo, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, changing hairstyle, face morphing, unstable face, background shift, glitching cuts, disappearing props; extra people, duplicated body, duplicated crane, more than three cranes, fewer than three cranes, broken continuity, wrong lip sync; modern objects, modern architecture, modern clothing details, electronic sounds, generated subtitles, unrealistic bird motion.

Original prompt

诗意武侠 × 冷面反转喜剧|白鹤认米不认琴
【风格】
电影级写实质感,真人写实。
融合诗意武侠的克制美学、冷面观察式喜剧、优雅的古典中国构图、傍晚柔和逆光、低饱和玉绿色与象牙白色调、细腻胶片颗粒,以及可控的灵鹤群体调度。
反转核心是:动物本能轻易击败仙人威严。
前半段营造高雅、庄重、仙气十足的“琴音引鹤”奇观,后半段用一个极小的日常意外完成冷面拆台。
全片为纯古代中国世界,不出现任何现代元素。
【生成重点】
专为 Seedance 2.0 的以下能力优化:
参考图身份锁定
多镜头连续叙事
双角色稳定
群体动物调度
原生音画同步
【时长】
严格 10 秒

【画幅】
16:9 横屏,24fps

【镜头结构】
三个干净清晰的镜头

一、角色锁定

角色 A|剑仙师姐

使用 @图片 1 严格锁定:

面部
发型

身材比例
服装

鞋子
饰品

人物设定:
25–30 岁东亚女性
椭圆脸

白皙自然肤色

深色杏眼

黑色长发半挽,以白玉簪固定

身材高挑纤细

固定服装:

白色刺绣真丝汉服

分层半透明宽袖

银色腰封

玉佩

白色布靴

人物气质:

清冷

从容

端庄

自信

极力维持仙人威严

角色 B|小师妹

使用 @图片 2 严格锁定身份:

20–25 岁东亚女性

圆润灵动的脸

黑发编辫

身形娇小

固定服装:

青绿色亚麻汉服

深色布腰带

木簪
黑色布鞋

固定道具:

手中拿着一个装谷物的小竹篮

人物气质:

自然

真诚

崇拜师姐

补刀时语气平静,冷面感强

二、动物锁定

白鹤群

全片 始终只出现 3 只白鹤。
不能变成 2 只、4 只或更多。
三只白鹤必须保持:

白色羽毛

修长体态

灵鹤气质

动作自然真实

群体调度清楚

前半段它们动作整齐优雅。
后半段它们会因为谷物立刻脱离庄严状态,变得本能、急切、略显嘈杂。

三、场景设定
【场景】
云海之上的宗门临崖石台。
场景中固定包含:
古松
石栏
木质凉亭

青铜香炉

随风飘动的祈福布条

石板地面

云海与远山雾气

一张古琴

空间关系要求明确稳定:
剑仙师姐端坐在古琴前

小师妹站在凉亭旁
白鹤最初在上空盘旋,后降落到师姐身后
谷物最后洒落在古琴与小师妹之间的石板地面上
运镜要让观众始终清楚知道:

古琴的位置

师姐的位置

小师妹的位置

三只白鹤的位置

谷物洒落的位置

四、分镜设计

【镜头 1|0–3s】

低角度全景横向跟拍

云海之上的宗门临崖石台,古松、石栏、木质凉亭、青铜香炉和祈福布条共同建立一个清楚稳定的古典空间。
同一个穿白色刺绣真丝汉服的剑仙师姐端坐于古琴前,从容抚琴。
同样 3 只白鹤在上空整齐盘旋一圈。
同一个穿青绿色汉服的小师妹站在凉亭旁等待,手中拿着同一个装谷物的小竹篮。

摄影机低角度横向缓慢移动,建立人、琴、鹤、凉亭和石台之间的稳定关系。
山风自然吹动头发、宽袖、祈福布条和云雾。

【声音】

古琴声

山风声

远处鹤鸣

云海环境声

轻微衣料摩擦声

【本镜头主要动作】
师姐抚琴,三只白鹤在上空整齐盘旋。

【镜头 2|3–6s】

中景侧面镜头

同一个剑仙师姐闭上双眼,弹完一段优雅琴句,淡然说道:
“琴音一起,百鸟来朝。”
同样 3 只白鹤以优美整齐的队形缓缓降落在她身后,动作同步,翅膀整齐收拢,场面庄重而好看。
同一个小师妹露出发自内心的崇拜神情。
背景始终保持同一片石台、同一座凉亭、同一组石栏、同一尊香炉和同一片云海。
镜头要拍清:
师姐在前
白鹤落在她身后
小师妹站在一侧观看

【声音】
优雅古琴尾音

白鹤振翅声

鹤鸣声

师姐原生普通话台词,口型准确

【本镜头主要动作】
三只白鹤整齐降落在师姐身后。

【镜头 3|6–10s】

近景跟随揭晓,转极近特写

同一个小师妹手中的同一个竹篮不小心轻轻倾斜,一小股谷粒洒落到石板地面上。
谷粒落地声音清晰。
同样 3 只白鹤瞬间被吸引,立刻放弃庄严整齐的队形,争先恐后地从剑仙师姐身边冲过去,低头啄食谷粒。
动作变得急切、直接、充满动物本能。
其中 一只白鹤 踩上同一张古琴,意外拨出一声刺耳乱音。
另外两只白鹤在石板上快速啄食。

同一个小师妹面无表情地说道:
“师姐,它们认的是米。”
随后切到极近特写:
同一个剑仙师姐双手悬停在琴弦上方,身体完全僵住。
一侧眉毛轻轻抽动。
一根松散的白色羽毛缓缓落在她的白玉簪上。
她没有发作,只是沉默僵住,体面正在一点点崩掉。

【声音】

谷粒散落声
白鹤急促脚步声

啄食声
振翅扑动声

古琴被踩出的刺耳乱音

小师妹原生普通话台词

结尾留一点尴尬安静

【本镜头主要动作】
谷粒洒落后,三只白鹤瞬间冲去吃米,拆穿“琴音引鹤”的庄严感。

五、最后 4 秒动作顺序锁定

为提高 Seedance 2.0 的执行稳定性,最后 4 秒严格按以下顺序发生:

小师妹手中的竹篮轻微倾斜

一小股谷粒洒落到石板地面

三只白鹤同时被谷粒吸引
三只白鹤放弃庄严队形,快速冲向谷粒

其中一只白鹤踩上古琴,拨出刺耳乱音
小师妹平静说出台词
切极近特写
剑仙师姐僵住,眉毛轻抽

一根白羽毛落到她的玉簪上
不要把这些动作同时挤在一起,不要打乱顺序。
六、表演要求
剑仙师姐

前 6 秒必须真正高雅、庄重、从容

台词要像仙门高人陈述事实

反转后不能大喊或夸张崩坏

喜剧点来自“威严被无声击穿”

重点表现:手悬停、身体僵住、眉毛轻抽、沉默破防
小师妹
前半段真心崇拜

洒落谷物后不要慌张

补刀语气平静、自然、像客观总结

她不需要夸张笑,只要冷静一句话就够

三只白鹤

前半段优雅整齐,像被琴音召来的灵禽

后半段立刻回归动物本能

这种反差是主要笑点来源之一

七、物理与连续性要求

物理效果
白鹤振翅、落地、冲刺、啄食动作自然真实

真丝衣料、头发、羽毛、云雾、布条运动自然
谷粒落地要有清楚的小颗粒感

古琴被踩时要有真实触弦反馈

羽毛落在玉簪上的动作要轻缓可信

连续性要求

全片始终只出现 3 只白鹤

不增加额外人物

不复制白鹤

不复制身体或道具

人物面部、服装、发型、身材比例全程稳定

古琴、石台、凉亭、香炉、石栏、竹篮位置保持一致

镜头衔接自然,不瞬移,不背景漂移

八、声音要求
原生同步普通话对白
古琴声、振翅声、鹤鸣、谷粒散落声、石板脚步声都要清楚同步

不生成字幕

不出现现代音效

整体声音要有电影感 and 真实环境层次

九、技术要求

严格总时长 10 秒
三个干净清晰的镜头

16:9 横屏

24fps

Seedance 2.0 参考图锁定稳定

多镜头叙事连续
双角色与三只白鹤关系明确

普通话口型准确

不增加额外人物

不生成字幕
十、负面词
blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, subtitles, captions, logo, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, changing hairstyle, face morphing, unstable face, background shift, glitching cuts, disappearing props; extra people, duplicated body, duplicated crane, more than three cranes, fewer than three cranes, broken continuity, wrong lip sync; modern objects, modern architecture, modern clothing details, electronic sounds, generated subtitles, unrealistic bird motion
Generate this with Seedance 2 on VdoBloom.
Opens the image-to-video creator with this prompt and Seedance 2 pre-selected.
✨ Generate Video