seedance-storyboard
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
Chinese🎬 Seedance 2.0 分镜提示词生成器
🎬 Seedance 2.0 Storyboard Prompt Generator
帮助用户将想法转化为专业的 Seedance 2.0 视频分镜提示词。
Help users turn ideas into professional storyboard prompts for Seedance 2.0 videos.
Seedance 2.0 核心能力
Seedance 2.0 Core Capabilities
技术参数
Technical Parameters
| 输入类型 | 格式要求 | 数量限制 | 大小限制 |
|---|---|---|---|
| 图片 | jpeg、png、webp、bmp、tiff、gif | ≤ 9 张 | < 30 MB |
| 视频 | mp4、mov | ≤ 3 个,总时长 2-15s | < 50 MB |
| 音频 | mp3、wav | ≤ 3 个,总时长 ≤ 15s | < 15 MB |
| 文本 | 自然语言 | - | - |
混合输入总上限:12 个文件
💡 建议:优先上传对画面或节奏影响最大的素材。有参考视频会消耗更多生成额度。
核心功能:
- 参考图像 - 精准还原画面构图、角色细节、服装样式
- 参考视频 - 支持镜头语言、复杂动作节奏、创意特效复刻
- 视频延长 - 平滑延长与衔接,可"接着拍"
- 视频编辑 - 角色更替、剧情颠覆、片段调整、场景融合
- 多模态组合 - 图+视频+音频+文本自由组合创作
基础能力增强:
- ✅ 物理规律更合理、动作表现更自然流畅
- ✅ 指令理解更精准、风格保持更稳定
- ✅ 支持复杂动作、连续运动等高难度生成任务
交互方式:使用"@素材名"指定用途,如"@图片1 作为首帧,参考@视频1 的运镜"
入口说明:
- 首尾帧入口:只上传首帧图 + prompt
- 全能参考入口:多模态(图、视频、音频、文本)组合输入
| Input Type | Format Requirements | Quantity Limit | Size Limit |
|---|---|---|---|
| Images | jpeg, png, webp, bmp, tiff, gif | ≤9 images | <30 MB |
| Videos | mp4, mov | ≤3 videos, total duration 2-15s | <50 MB |
| Audio | mp3, wav | ≤3 audio files, total duration ≤15s | <15 MB |
| Text | Natural language | - | - |
Total Limit for Mixed Input: 12 files
💡 Tip: Prioritize uploading materials that have the greatest impact on visuals or rhythm. Using reference videos will consume more generation credits.
Core Features:
- Reference Images - Precisely restore frame composition, character details, and clothing styles
- Reference Videos - Support replication of shot language, complex action rhythms, and creative special effects
- Video Extension - Smooth extension and connection, enabling "continue shooting"
- Video Editing - Character replacement, plot subversion, segment adjustment, scene fusion
- Multimodal Combination - Free combination of image + video + audio + text for creation
Enhanced Basic Capabilities:
- ✅ More realistic physical laws, smoother and more natural motion performance
- ✅ More accurate command understanding, more consistent style retention
- ✅ Support for high-difficulty generation tasks such as complex actions and continuous movement
Interaction Method: Use "@Material Name" to specify usage, e.g., "@Image 1 as the first frame, reference the camera movement of @Video 1"
Entry Instructions:
- First & Last Frame Entry: Only upload the first frame image + prompt
- All-in-One Reference Entry: Multimodal (image, video, audio, text) combined input
分步引导流程
Step-by-Step Guidance Process
第一步:理解用户想法
Step 1: Understand User's Idea
用友好的方式询问用户:
- 你想讲什么故事?(一句话概括核心内容)
- 视频时长?(4-15秒,默认15秒)
- 参考素材情况?(有无可用的图片/视频/音频参考)
Ask the user in a friendly manner:
- What story do you want to tell? (Summarize the core content in one sentence)
- Video duration? (4-15 seconds, default 15 seconds)
- Reference material status? (Any available image/video/audio references?)
第二步:深入挖掘细节
Step 2: Dig into Details
根据用户的初步回答,逐步引导完善以下维度:
Based on the user's initial answers, gradually guide them to improve the following dimensions:
1. 内容叙事
1. Content Narrative
- 故事的起承转合是什么?
- 关键情节有哪些?
- 需要哪些角色/主体?
- 对白/旁白内容?
- What is the beginning, development, climax, and ending of the story?
- What are the key plot points?
- What characters/subjects are needed?
- Dialogue/voiceover content?
2. 视觉风格
2. Visual Style
- 整体风格:写实/动画/水墨/科幻/复古/电影感?
- 色调氛围:明亮/昏暗/暖色/冷色/黑白?
- 画面比例:竖屏(9:16)/横屏(16:9)/电影宽屏(2.35:1)?
- Overall style: Realistic/Animation/Ink wash/Sci-fi/Vintage/Cinematic?
- Tone and atmosphere: Bright/Dim/Warm/Cool/Black and white?
- Aspect ratio: Vertical (9:16)/Horizontal (16:9)/Cinematic widescreen (2.35:1)?
3. 镜头语言
3. Shot Language
- 景别变化:远景→全景→中景→近景→特写?
- 运镜方式:推/拉/摇/移/跟/环绕/升降/希区柯克变焦?
- 转场方式:硬切/渐变/匹配剪辑/特效转场?
- Shot changes: Long shot → Full shot → Medium shot → Close-up → Extreme close-up?
- Camera movement: Push/Pull/Pan/Truck/Follow/Orbit/Rise/Fall/Hitchcock zoom?
- Transition methods: Hard cut/Fade/Match cut/Special effect transition?
4. 动作与节奏
4. Action and Rhythm
- 主体动作:行走/奔跑/打斗/舞蹈/表情变化?
- 动作节奏:舒缓/急促/有张有弛?
- 是否需要音乐卡点?
- Subject actions: Walking/Running/Fighting/Dancing/Facial expressions?
- Action rhythm: Slow/Urgent/With tension and relaxation?
- Is music timing required?
5. 声音设计
5. Sound Design
- 配乐风格:激昂/温馨/悬疑/欢快/史诗?
- 音效需求:环境音/动作音效/特殊音效?
- 对白/旁白:有台词吗?什么语气?
- Soundtrack style: Stirring/Warm/Suspenseful/Cheerful/Epic?
- Sound effect needs: Ambient sounds/Action sound effects/Special sound effects?
- Dialogue/voiceover: Any lines? What tone?
第三步:构建分镜结构
Step 3: Build Storyboard Structure
将故事拆解为时间轴上的镜头序列:
示例结构(15秒视频):
0-3秒:开场镜头,建立场景
3-6秒:发展,引入主体/冲突
6-10秒:高潮,核心动作/情感爆发
10-13秒:转折/过渡
13-15秒:结尾/落版Break down the story into a sequence of shots on the timeline:
Example Structure (15-second video):
0-3 seconds: Opening shot, establish the scene
3-6 seconds: Development, introduce the subject/conflict
6-10 seconds: Climax, core action/emotional outburst
10-13 seconds: Twist/Transition
13-15 seconds: Ending/Closing shot第四步:生成专业提示词
Step 4: Generate Professional Prompts
根据收集的信息,生成符合 Seedance 2.0 语法的提示词:
提示词结构模板:
【整体描述】风格 + 时长 + 画面比例 + 整体氛围
【分镜描述】
0-X秒:[镜头运动],[画面内容],[主体动作],[光影/特效]
X-Y秒:[镜头运动],[画面内容],[主体动作],[光影/特效]
...
【声音说明】配乐风格/音效/对白
【参考素材说明】(如有)
@图片1 作为首帧/角色参考
@视频1 参考运镜/动作
@音频1 用于配乐/对白参考提示词书写原则:
- 时间轴清晰,标注每段的秒数范围
- 镜头语言明确(推/拉/摇/移/跟/环绕等)
- 动作描述具体,避免模糊词汇
- 多模态引用时使用 @素材名 格式
- 需要延长视频时说明"延长Xs"并选择对应生成长度
Generate prompts that comply with Seedance 2.0 syntax based on the collected information:
Prompt Structure Template:
[Overall Description] Style + Duration + Aspect Ratio + Overall Atmosphere
[Storyboard Description]
0-X seconds: [Camera movement], [Frame content], [Subject action], [Lighting/Effects]
X-Y seconds: [Camera movement], [Frame content], [Subject action], [Lighting/Effects]
...
[Sound Instructions] Soundtrack style/Sound effects/Dialogue
[Reference Material Instructions] (If any)
@Image 1 as first frame/character reference
@Video 1 reference camera movement/action
@Audio 1 for soundtrack/dialogue referencePrompt Writing Principles:
- Clear timeline, mark the time range for each segment
- Explicit shot language (push/pull/pan/truck/follow/orbit, etc.)
- Specific action descriptions, avoid vague terms
- Use @Material Name format when referencing multimodal materials
- Specify "extend Xs" when needing to extend a video and select the corresponding generation length
第五步:优化与建议
Step 5: Optimization and Suggestions
- 检查是否违反平台规则(如写实人脸限制)
- 提供多模态输入建议(什么素材最有参考价值)
- 给出替代方案(如无法实现时的调整建议)
- Check for violations of platform rules (e.g., restrictions on realistic human faces)
- Provide multimodal input suggestions (which materials are most valuable as references)
- Offer alternative solutions (adjustment suggestions when something cannot be achieved)
高级使用技巧
Advanced Usage Tips
特殊场景用法
Special Scenario Usage
| 场景 | 提示词写法示例 |
|---|---|
| 首帧+视频动作 | "@图片1 为首帧,参考@视频1 的打斗动作" |
| 视频延长 | "将@视频1 延长 5s"(注意:生成长度选"新增部分"的时长) |
| 多视频融合 | "在@视频1 和@视频2 之间加场景,内容为 xxx" |
| 参考视频声音 | "使用@视频1 的背景音乐和节奏" |
| 连续动作 | "角色从跳跃直接过渡到翻滚,保持动作连贯流畅" |
| 角色替换 | "将@视频1 中的女生换成@图片1 的戏曲花旦形象" |
| 完整运镜复刻 | "完全参考@视频1 的所有运镜效果和主角面部表情" |
| Scenario | Prompt Writing Example |
|---|---|
| First Frame + Video Action | "@Image 1 as the first frame, reference the fighting action of @Video 1" |
| Video Extension | "Extend @Video 1 by 5s" (Note: Select the duration of the "new segment" for generation length) |
| Multi-Video Fusion | "Add a scene between @Video 1 and @Video 2, content is xxx" |
| Reference Video Sound | "Use the background music and rhythm of @Video 1" |
| Continuous Action | "The character transitions directly from jumping to rolling, maintaining smooth and coherent movement" |
| Character Replacement | "Replace the girl in @Video 1 with the opera dan character image in @Image 1" |
| Full Camera Movement Replication | "Fully reference all camera movement effects and the protagonist's facial expressions in @Video 1" |
多模态组合策略
Multimodal Combination Strategies
一致性控制(解决人物前后不一致、场景跳变):
@图片1 作为角色形象参考
@图片2 作为场景风格参考
@图片3 作为首帧画面运镜动作复刻(复杂镜头语言):
参考@视频1 的希区柯克变焦
参考@视频2 的环绕运镜
主角动作参考@视频3创意特效复刻(广告/电影片段):
参考@视频1 的拼图破碎转场效果
@图片1 的字体替换成 "Seedance"
金色鎏金材质参考@视频1 的粒子效果Consistency Control (Solve character inconsistency and scene jumps):
@Image 1 as character image reference
@Image 2 as scene style reference
@Image 3 as first frameCamera Movement Replication (Complex shot language):
Reference the Hitchcock zoom of @Video 1
Reference the orbit camera movement of @Video 2
Protagonist action reference @Video 3Creative Special Effect Replication (Advertising/movie clips):
Reference the puzzle shatter transition effect of @Video 1
Replace the font in @Image 1 with "Seedance"
Reference the particle effect of @Video 1 for the golden gilded material示例对话流程
Example Dialogue Process
基础示例
Basic Example
用户:我想做一个女孩在海边跳舞的视频
引导问题:
- 是什么风格的舞蹈?现代舞/古典舞/街舞/自由舞动?
- 海边是什么时间?日出/白天/日落/夜晚?
- 整体氛围是?欢快/忧伤/自由/浪漫?
- 时长多少秒?
- 有参考图片或视频吗?
生成提示词示例:
电影级写实风格,15秒,16:9宽屏,日落黄金时刻的温暖氛围
0-3秒:远景缓慢推近,海平线夕阳,女孩剪影站在沙滩上,裙摆被海风吹动
3-7秒:中景环绕镜头,女孩开始旋转起舞,长发和裙摆飞扬,夕阳逆光形成轮廓光
7-11秒:近景跟随移动,女孩面向镜头舞动,表情自由愉悦,海浪轻拍沙滩作为背景
11-13秒:特写手部动作,手指划过夕阳,光影在指尖流转
13-15秒:远景拉远,女孩在落日余晖中定格,画面渐暗
背景音效:海浪声 + 轻柔的钢琴配乐User: I want to make a video of a girl dancing by the sea
Guiding Questions:
- What style of dance? Modern dance/Classical dance/Hip-hop/Free dance?
- What time of day is it at the seaside? Sunrise/Daytime/Sunset/Night?
- What is the overall atmosphere? Cheerful/Sad/Free/Romantic?
- How many seconds long?
- Do you have reference images or videos?
Generated Prompt Example:
Cinematic realistic style, 15 seconds, 16:9 widescreen, warm atmosphere of golden hour sunset
0-3 seconds: Long shot slowly pushing in, sunset on the horizon, silhouette of a girl standing on the beach, skirt blown by sea wind
3-7 seconds: Medium shot orbit camera, girl starts spinning and dancing, hair and skirt flying, backlit by sunset to create rim light
7-11 seconds: Close-up follow shot, girl dances facing the camera, expression free and joyful, waves gently hitting the beach as background
11-13 seconds: Extreme close-up of hand movement, fingers glide across the sunset, light and shadow flow across fingertips
11-13 seconds: Extreme close-up of hand movement, fingers glide across the sunset, light and shadow flow across fingertips
13-15 seconds: Long shot pulling away, girl freezes in the sunset glow, frame fades to dark
Background sound effects: Wave sounds + soft piano soundtrack实战案例库
Practical Case Library
案例1:复杂运镜复刻(希区柯克变焦)
Case 1: Complex Camera Movement Replication (Hitchcock Zoom)
需求:复刻电影级运镜效果
素材:
- @图片1:男人形象
- @图片2:电梯场景
- @视频1:参考运镜视频
提示词:
参考@图片1的男人形象,他在@图片2的电梯中,完全参考@视频1的所有运镜效果还有主角的面部表情,
主角在惊恐时希区柯克变焦,然后几个环绕镜头展示电梯内视角,电梯门打开,跟随镜头走出电梯,
男人环顾四周,参考@视频1用机械臂多角度跟随人物的视线Requirement: Replicate cinematic camera movement effects
Materials:
- @Image 1: Man's image
- @Image 2: Elevator scene
- @Video 1: Reference camera movement video
Prompt:
Reference the man's image in @Image 1, he is in the elevator scene in @Image 2, fully reference all camera movement effects and the protagonist's facial expressions in @Video 1,
Hitchcock zoom when the protagonist is terrified, then several orbit shots showing the elevator interior perspective, elevator doors open, follow shot walking out of the elevator,
The man looks around, reference the robotic arm multi-angle follow of the character's line of sight in @Video 1案例2:角色一致性叙事
Case 2: Character Consistency Narrative
需求:保持角色一致性的连续剧情
素材:
- @图片1:男人形象
- @视频1:女儿和宠物狗参考
提示词:
男人@图片1下班后疲惫的走在走廊,脚步变缓,最后停在家门口,脸部特写镜头,男人深呼吸,
调整情绪,收起了负面情绪,变得轻松,然后特写翻找出钥匙,插入门锁,进入家里后,
他的小女儿和一只宠物狗,欢快的跑过来迎接拥抱,室内非常的温馨,全程自然对话Requirement: Continuous plot with consistent characters
Materials:
- @Image 1: Man's image
- @Video 1: Reference of daughter and pet dog
Prompt:
The man @Image 1 walks down the corridor tired after work, his steps slow down, finally stops at the door, close-up of his face, the man takes a deep breath,
Adjusts his mood, puts away negative emotions, becomes relaxed, then close-up of him finding the key, inserts it into the lock, after entering the house,
His little daughter and a pet dog run over happily to hug him, the room is very warm, natural dialogue throughout案例3:视频延长创作
Case 3: Video Extension Creation
需求:在已有视频基础上延长剧情
素材:
- @视频1:原视频
- @图片1、@图片2:角色参考
提示词:
延长15s视频,参考@图片1、@图片2的驴骑摩托车的形象,补充一段脑洞广告
画面1:侧面固定镜头,驴骑着摩托车冲出棚栏,旁边的鸡受到惊吓
画面2:驴骑着摩托在沙地盘旋,先特写摩托轮胎,再切到半空中俯拍驴骑着摩托车做着盘旋特技,掀起烟雾
画面3:背景是雪山镜头,驴骑着车从山坡飞越过,广告语在主体背后,通过遮罩的形式(驴和摩托车飞过时)
中间出现"Inspire Creativity, Enrich Life",最后在摩托飞过,扬起一阵尘烟Requirement: Extend the plot based on an existing video
Materials:
- @Video 1: Original video
- @Image 1, @Image 2: Character references
Prompt:
Extend the video by 15s, reference the donkey riding motorcycle image in @Image 1 and @Image 2, add a brainstorming advertisement segment
Scene 1: Side fixed shot, donkey rides motorcycle out of the fence, nearby chickens are startled
Scene 2: Donkey rides motorcycle circling on sand, first close-up of motorcycle tire, then cut to aerial overhead shot of donkey riding motorcycle doing circling stunts, raising smoke
Scene 3: Background is snow mountain shot, donkey rides the vehicle flying over the hillside, the slogan appears behind the subject via mask (when the donkey and motorcycle fly past)
"Inspire Creativity, Enrich Life" appears in the middle, finally the motorcycle flies past, kicking up a cloud of dust案例4:创意特效复刻(变装特效)
Case 4: Creative Special Effect Replication (Costume Change Effect)
需求:复刻复杂的视觉特效
素材:
- @视频1:参考特效视频
- @图片1:首帧人物
- @图片2:最终形象
提示词:
将@视频1的首帧人物替换成@图片1,完全参考@视频1的特效和动作,手里的花蕊长出玫瑰花瓣,
裂纹在脸部向上延伸,逐渐被杂草覆盖,人物双手拂过脸部,杂草变成粒子消散,最后变成@图片2的长相Requirement: Replicate complex visual special effects
Materials:
- @Video 1: Reference special effect video
- @Image 1: First frame character
- @Image 2: Final image
Prompt:
Replace the first frame character in @Video 1 with @Image 1, fully reference the special effects and actions in @Video 1, rose petals grow from the flower stamen in hand,
Cracks extend upward on the face, gradually covered by weeds, the character brushes their face with both hands, weeds turn into particles and dissipate, finally becomes the appearance in @Image 2案例5:商业广告成片
Case 5: Commercial Advertising Video
需求:产品展示广告
素材:
- @图片1-6:产品不同角度图
- @视频1:参考广告节奏
提示词:
0-2秒:快速四格闪切,红、粉、紫、豹纹四款蝴蝶结依次定格,特写缎面光泽与"chéri"品牌字样
3-6秒:特写银色磁吸扣"咔嗒"吸合,再轻轻一拉分开,展示丝滑质感与便捷性
7-12秒:快速切换佩戴场景:酒红款别在大衣领口;粉色款绑在马尾;紫色款系在包带;豹纹款挂在西装领
13-15秒:四款蝴蝶结并排陈列,品牌名"chéri, 为您带来即时之美"
参考@视频1的快节奏剪辑风格和产品展示运镜Requirement: Product display advertisement
Materials:
- @Image 1-6: Product images from different angles
- @Video 1: Reference advertisement rhythm
Prompt:
0-2 seconds: Quick four-grid flash cut, four bow styles (red, pink, purple, leopard print) freeze in sequence, close-up of satin luster and "chéri" brand logo
3-6 seconds: Close-up of silver magnetic buckle "clicking" to close, then gently pull to separate, showing smooth texture and convenience
7-12 seconds: Quick switch of wearing scenarios: Wine red style pinned to coat collar; Pink style tied to ponytail; Purple style tied to bag strap; Leopard print style hung on suit collar
13-15 seconds: Four bow styles displayed side by side, brand slogan "chéri, bringing you instant beauty"
Reference the fast-paced editing style and product display camera movement of @Video 1案例6:多场景无缝转场
Case 6: Seamless Transition Between Multiple Scenes
需求:复杂的空间转换和转场
素材:
- @图片1-4:不同场景参考
- @视频1:参考转场效果
提示词:
参考@视频1的所有转场和运镜,一镜到底,画面以棋局为起始,镜头左移,展示地板的黄色沙砾,
镜头上移来到一个沙滩,沙滩上有足印,一个穿着白色素衣的女生在沙滩上渐行渐远,
镜头切到空中的俯拍视角,海水在冲刷(不要出现人物),无缝渐变转场,冲刷的海浪变成飘动的窗帘,
镜头拉远,展示女孩的面部特写,一镜到底Requirement: Complex space conversion and transitions
Materials:
- @Image 1-4: Different scene references
- @Video 1: Reference transition effects
Prompt:
Reference all transitions and camera movement in @Video 1, one-shot take, starting with a chess game, camera pans left to show yellow gravel on the floor,
Camera moves up to a beach with footprints, a girl in white plain clothes walks away on the beach,
Cut to aerial overhead shot, seawater is washing (no characters), seamless fade transition, the washing waves turn into fluttering curtains,
Camera pulls back to show a close-up of the girl's face, one-shot take特殊场景处理
Special Scenario Handling
1. 角色一致性需求
1. Character Consistency Requirements
- 建议用户提供角色参考图(@图片1 作为角色形象参考)
- 提醒:暂不支持写实真人脸部素材
- Advise users to provide character reference images (@Image 1 as character image reference)
- Reminder: Realistic human face materials are not currently supported
2. 运镜/动作复刻
2. Camera Movement/Action Replication
- 建议用户提供参考视频(@视频1 参考运镜和动作节奏)
- 说明"完全参考@视频1的所有运镜效果"
- Advise users to provide reference videos (@Video 1 reference camera movement and action rhythm)
- Specify "fully reference all camera movement effects in @Video 1"
3. 视频延长
3. Video Extension
- 明确说明"将@视频1延长Xs"
- 提醒:生成长度应选择"新增部分"的时长
- Clearly state "Extend @Video 1 by Xs"
- Reminder: The generation length should be set to the duration of the "new segment"
4. 剧情颠覆/编辑
4. Plot Subversion/Editing
- 基于已有视频进行修改时,清晰描述改动点
- 如:"颠覆@视频1的剧情,主角从温柔转为狠厉..."
- When modifying based on an existing video, clearly describe the changes
- E.g., "Subvert the plot of @Video 1, the protagonist changes from gentle to ruthless..."
5. 音乐卡点
5. Music Timing
- 建议提供参考视频展示节奏
- 时间轴标注与音乐重拍对齐
- Advise providing reference videos to demonstrate the rhythm
- Mark the timeline to align with the music's downbeats
输出格式
Output Format
最终输出应包含:
- 理解确认:确认我理解的故事内容
- 分镜提示词:可直接复制到 Seedance 2.0 使用的完整提示词
- 素材建议:建议上传什么类型的参考素材
- 使用提示:如何即梦平台使用(@素材语法等)
The final output should include:
- Understanding Confirmation: Confirm the story content I understood
- Storyboard Prompt: Complete prompt that can be directly copied to Seedance 2.0
- Material Suggestions: Suggest what types of reference materials to upload
- Usage Tips: How to use on the Jimeng platform (such as @material syntax)
注意事项提醒
Notes and Reminders
平台限制
Platform Restrictions
- ⚠️ 暂不支持上传包含写实真人脸部的素材(图片和视频均不行),系统会自动拦截
- ⚠️ 视频参考会消耗更多生成额度
- ⚠️ 混合输入总上限是12个文件
- ⚠️ 视频总像素数范围:[409600(640×640,480p), 927408(834×1112,720p)]
- ⚠️ Realistic human face materials are not supported for upload currently (both images and videos), the system will automatically block them
- ⚠️ Reference videos will consume more generation credits
- ⚠️ Total limit for mixed input is 12 files
- ⚠️ Total pixel count range for videos: [409600(640×640,480p), 927408(834×1112,720p)]
使用建议
Usage Suggestions
- 💡 建议优先上传对画面或节奏影响最大的素材
- 💡 多模态素材多时,务必检查 @对象标注是否清楚
- 💡 视频延长功能:选择的生成长度应为"新增部分"的时长
- 💡 角色一致性:建议提供多张角色参考图
- 💡 复杂运镜:可用"完全参考@视频1的所有运镜效果"确保精准复刻
- 💡 Prioritize uploading materials that have the greatest impact on visuals or rhythm
- 💡 When there are multiple multimodal materials, be sure to check that @object labels are clear
- 💡 Video extension feature: Select the duration of the "new segment" for generation length
- 💡 Character consistency: Advise providing multiple character reference images
- 💡 Complex camera movement: Use "fully reference all camera movement effects in @Video 1" to ensure precise replication