LogoAhaPrompt
  • 提示詞
  • 精選
  • 模型
  • AI 反推
  • 定價
  • 投稿
  • 部落格
AhaPrompt

圖片、影片、音訊 AI 提示詞庫

探索

  • 全部提示詞
  • 精選提示詞
  • 圖片提示詞
  • 影片提示詞
  • 音訊提示詞
  • AI 模型

工具

  • AI 反推工作臺
  • 圖片反推提示詞
  • 影片反推提示詞

資源

  • 定價方案
  • 投稿提示詞
  • 部落格

版權所有 © 2026 AhaPrompt 保留所有權利。

隱私政策條款與條件退款政策版權申訴 / DMCA
首頁AI 提示詞庫AI 影片提示詞参考照片角色转面图

参考照片角色转面图

ByteDanceSeedance 2.5ByteDancevideo

参考照片中人物的角色转面图,六个视图:全身正面、侧面和背面 + 特写正面、侧面和背面。严格保留参考中100%的面部、头发、身体、服装及所有细节。 干净的灰色背景,一致且自然的光照。

詳細資訊
类型
AI 影片提示詞
畫面比例
9:16
Duration
0:11
更新於
2026年10月31日
分類
CharacterDigital ArtPortrait
Source

如何使用這條提示詞

  1. 1點選上方複製按鈕,複製完整提示詞。
  2. 2開啟 ByteDance 的 Seedance 2.5,或你常用的 AI 影片生成工具。
  3. 3貼上提示詞,保留鏡頭與運動描述詞——它們決定了畫面的運鏡效果。
  4. 4按示例參數(9:16 · 0:11 · Character · Digital Art · Portrait)生成,再小步迭代——措辭的細微變化會顯著影響結果。

常見問題

這段影片是用哪個 AI 模型生成的?

本片段由 ByteDance 的 Seedance 2.5 生成,頁面上的提示詞就是生成它的原始文字。

這條提示詞用了哪些參數?

示例由 Seedance 2.5 生成,參數:9:16 · 0:11 · Character · Digital Art · Portrait。沿用這些參數可獲得最接近的效果。

怎樣才能生成相似的效果?

先原樣複製提示詞,只改主體、保留鏡頭/運動/光影描述。影片模型對措辭很敏感,多試幾次為佳。

可以修改這條提示詞嗎?

當然可以。提示詞最適合作為起點——替換場景、調整節奏詞或寬高比,適配你的專案。

更多 Seedance 2.5 提示詞

0:30
ByteDanceSeedance 2.5
GIRL = MAIN CHARACTER: Young Korean woman in her early 20s, natural makeup, black wavy hair tied into a messy side ponytail, loose peach T-shirt, white pajama pants, black sneakers and simple necklace. Maintain exact identity and appearance. LOCATION: Small Korean apartment courtyard during a warm windy afternoon. Concrete walls, laundry lines, potted plants, bicycles and neighboring homes. CAMERA / LOOK: Early-2000s consumer DV camcorder. Strong handheld shake, autofocus hunting, exposure pumping, faded colors, soft contrast, motion blur and imperfect zooms. SEQUENCE: She hangs freshly washed clothes on a long outdoor clothesline. Everything looks calm until a sudden strong gust sends one shirt flying off the line. She runs after it while laughing. The camera operator follows shakily. The shirt lands on a parked bicycle. She grabs it, tries to hang it again, and another piece of laundry immediately blows away. She looks toward the camera in disbelief before chasing it down the lane. AUDIO: Wind, fabric flapping, birds, footsteps, distant scooters and neighborhood voices. No music. Candid domestic humor, realistic fabric movement, nostalgic DV imperfections and believable everyday chaos.
檢視詳情
0:30
ByteDanceSeedance 2.5
【整体风格与核心故事】 严格30秒,16:9横屏,Seedance 2.5,真人电影级古风仙侠反转喜剧,原生普通话音画同步。 作品:《说好的合影呢?》 整体采用真实自然的中国古风影视美学,细腻肤质、真实丝绸衣料、自然体积光、丰富环境纵深和克制的人物表演。 借鉴经典视觉喜剧中的信息差、物体运用、行为重复与角色互换,以及现实生活中为了拍出一张满意的照片而不断折腾身边人的熟悉情境。 主线非常明确: 小师妹原本想和师姐一起留下一张好看的合影。 白衣师姐却过分在意自己在画面中是否完美。 她不断让小师妹举起铜镜、调整方向、寻找合适光线,逐渐把小师妹变成了自己的专属补光助手。 观众比小师妹更早意识到,这张所谓的“合影”实际上只有师姐一个人。 拍完以后,小师妹发现画面里只有自己举镜子的那只手。 师姐竟然还一本正经地解释“手拍得挺好”。 小师妹没有争吵,而是把铜镜交到师姐手里,让师姐亲自体验刚才的过程。 最后一个镜头完全呼应开头,只交换两个人的位置与身份。 核心喜剧来自: 认真得过分的师姐; 逐渐察觉问题的小师妹; 一个人追求出片,另一个人被迫承担全部辅助工作的生活反差; 以及最后精确对称的角色互换。 不加入反派寻仇、师父拆台、突然战斗、随机魔法或无关反转。 【人物与人物关系】 角色ID A:同一个@图片1剑仙师姐。 25–30岁成年东亚女性,保持参考图中完全相同的人脸身份、五官骨相、肤质、比例、气质和发型。 高挑纤细,黑色长发半挽,白玉簪固定,始终穿白色刺绣真丝汉服、分层半透明宽袖、银色腰封、玉佩及白色布靴,持唯一一柄银色长剑。 性格冷静、端庄、自信,平时看起来不在乎外表,实际上对自己的形象相当讲究。 面对留影这件事,她甚至像研究剑法一样认真安排角度、姿势与光线。 角色ID B:同一个@图片2小师妹。 20–25岁成年东亚女性,保持参考图中完全相同的人脸身份、五官、面部比例、编发、体型和气质。 身形娇小,五官自然圆润灵动,穿同一套青绿色亚麻汉服、深色腰带、木簪和黑色布鞋,持唯一一柄深色钢剑。 性格直率、善于观察,开始时愿意帮师姐完成拍摄。 当她明白自己被当成举镜子的助手以后,不发火、不撒泼,只以同样的方式让师姐体验一次。 本集只有这两名核心人物,不额外增加师父、敌人或其他主要角色。 关键道具为一面直径约30厘米的古代抛光铜镜,以及一枚稳定放在石台上的仙侠留影石。 铜镜依靠真实自然光反射补光,有实际重量,始终由人物双手或单手合理握持。 留影石只用短促、柔和的灵光完成一次静态留影,不出现现代相机、电子屏幕或复杂法术界面。 【当前地点与空间连续性】 本轮新上传的所有背景与地点参考图共同决定视频的实际物理世界。 先无声提炼并整合参考图中能够兼容的真实地形、主要建筑关系、空间尺度、材质、植被、水体、天气、云层、山雾、主光方向、色彩、倒影和空气纵深。 重新组合为本轮独有、合理、连续的实际场景。 在这些真实空间中选择一处适合留影、可以安全站立的观景平台或庭院石台;如果参考图没有相应平台,就使用参考世界里真实存在、可供人物活动的位置。 绝不能凭空替换成陌生的泛化仙侠场景。 整段故事发生在同一个位置。 留影石固定在一个稳定石座上,面向师姐原本所站的位置。 师姐最初站在画面中央。 小师妹站在画面边缘,举起铜镜,将自然光柔和地反射到师姐脸上。 小师妹身体大部分处于留影石取景范围之外,只有举着铜镜的手和铜镜边缘刚好进入画面。 这种真实的几何位置必须从开头一直保留到反转揭示,不能为了笑点突然改变。 背景始终自然鲜活: 风吹动头发和衣袖; 树叶随风轻摆; 已有水体持续流动; 山雾按照真实地势缓慢变化; 光影和倒影随着摄影机视角合理变化。 但这些环境运动保持叙事中立,不主动推动或改变剧情。 【0–4秒:开场钩子】 开场首先给出一个非常漂亮的电影级中远景构图。 同一个白衣师姐站在参考图形成的壮阔自然空间中。 她姿态优雅,腰侧银色长剑自然垂落,浅色真丝宽袖被微风轻轻吹动。 初看仿佛一张真正绝美、庄严的剑仙电影剧照。 然而摄影机立刻进行一次小幅度、精准的横向移动。 画面旁边突然露出同一个青衣小师妹。 小师妹正吃力地高举一面铜镜,手臂已经开始发酸。 她看向师姐: 小师妹:“师姐,说好的合影呢?” 师姐甚至没有转头,只看着留影石的位置: 师姐:“镜子再抬高点。” 原本庄严的剑仙画面,瞬间变成一场让人十分熟悉的拍照折腾现场。 【4–10秒:人物目标与铺垫】 保持同一位置、同一面铜镜以及两人的真实空间关系。 中景双人镜头。 小师妹把铜镜举得更高一点。 师姐认真观察反射到自己脸上的柔和光线,轻轻移动下巴,又整理一缕头发。 师姐:“往左一点。” 小师妹按照要求移动手臂。 师姐:“再高一点。” 小师妹再次照做。 镜子有实际重量,她抬着的手臂已经明显开始疲劳。 小师妹终于有些疑惑: 小师妹:“师姐,我也入画了吗?” 师姐看了一眼留影石的取景方向。 她已经注意到小师妹大部分身体并没有进入画面。 短暂停顿。 师姐仍然非常平静: 师姐:“入了,放心。” 小师妹暂时选择相信。 师姐脸上那一点极其细微的不自然,让观众开始意识到她心里其实很清楚。 【10–18秒:喜剧冲突升级】 仍然保持完全相同的地点、人物和道具关系。 师姐先稍微转动身体寻找更适合面部的光线,再调整银色长剑的角度,使自己的整体轮廓更加优雅。 她只改变两次姿态。 但每次调整,都需要小师妹重新改变铜镜角度,才能让反射光保持稳定。 镜头不使用快速蒙太奇,而是完整记录两次连续、合理的动作。 小师妹一边举着铜镜,一边尝试把脑袋伸入留影石的取景范围,希望自己也能出现在合影里。 师姐立即轻轻提醒: 师姐:“别晃,马上好了。” 小师妹只好停住。 她保持着一个越来越不舒服的举镜姿势。 此刻给出一个留影石实际取景方向的客观镜头。 观众可以清楚看见: 画面中央是姿态完美的白衣师姐。 画面边缘只露出小师妹举铜镜的手和一点铜镜边框。 小师妹的脸和身体完全不在画面里。 这一镜头必须让观众先于角色理解最终的笑点。 留影石只发出一次极其柔和的短促灵光,完成静态留影。 不产生夸张魔法。 【18–25秒:关键变化与反转】 留影结束。 师姐终于放松姿态: 师姐:“成了。” 小师妹立即放下已经发酸的手臂,走到留影石旁边查看刚才的画面。 摄影机先保持自然的双人中景,再短暂展示刚刚记录的构图。 留影结果与之前取景镜头完全一致: 师姐漂亮地站在中央。 小师妹只有一只举镜子的手出现在画面边缘。 不出现任何文字。 小师妹认真看了一眼。 停顿半拍。 小师妹:“我呢?” 师姐非常认真地伸手指向画面边缘那只手。 师姐:“这儿,手拍得挺好。” 小师妹缓慢转头看向师姐。 小师妹:“你管这叫合影?” 师姐原本自信的表情第一次出现明显迟疑。 她看看留影画面,再看看小师妹。 终于意识到自己的解释听起来实在有些不像话。 不要让她继续狡辩,也不要增加第三人说话。 这一段的笑点依靠清楚的视觉证据、师姐一本正经的回答和小师妹的表情成立。 【25–30秒:笑点回收与结尾】 小师妹没有继续争论。 她只是非常自然地把铜镜交到师姐手里。 小师妹:“行,换你举。” 师姐下意识接过铜镜。 小师妹已经走到刚才师姐所站的画面中央。 她轻轻整理青绿色衣袖,摆出一个自然舒展、确实很好看的古风姿态。 师姐此时只能站在画面旁边,像刚才小师妹那样举起铜镜。 摄影机重新回到几乎与开头一模一样的构图。 但这次发生了完全对称的角色互换: 中央是漂亮的青衣小师妹。 边缘是高举铜镜的白衣师姐。 小师妹看向师姐,用几乎完全相同的平静语气说道: 小师妹:“镜子再抬高点。” 师姐看向小师妹。 又看看自己举着的铜镜。 她终于明白,小师妹打算把刚才的待遇原样还给自己。 师姐轻轻叹了一口气。 只能认命地把铜镜再抬高一点。 在这个与开场严格呼应、两人地位完全交换的电影画面上精准结束。 不再增加师父补刀、不追加解释、不出现额外突发事件。 【演员表演与喜剧节奏】 师姐从头到尾都要认真对待这件小事。 她的笑点来自一个明明很在意自己形象,却始终表现得端庄得体的人。 小师妹从信任、配合,到怀疑、发现真相,最后冷静反击,情绪必须连续发展。 手臂疲劳逐步加重,但不使用夸张颤抖、摔倒或者卡通动作。 关键停顿包括: 师姐回答“入了,放心”之前的极短迟疑; 留影画面揭示之后小师妹的半拍沉默; 师姐说出“手拍得挺好”后的尴尬; 最后师姐被迫举镜时的无奈。 所有反应自然可信。 【摄影机与剪辑】 开场采用电影级优美中远景,再通过小幅横移揭示举镜的小师妹,让观众快速重新理解画面。 铺垫使用双人中景,清楚交代空间关系。 递进段落通过铜镜、手臂、剑身、脸部反光等适量细节镜头展现实际动作,但不使用连续碎切。 留影石取景镜头作为核心视觉信息,必须明确展示谁在画面里、谁在画面外。 反转通过人物查看留影结果和表情变化完成。 最后采用与开场相互呼应的对称构图,使观众在无需额外解释的情况下理解角色互换。 镜头变化自然,不环绕炫技,不随意切换环境。 【普通话对白与声音设计】 全部使用自然普通话。 实际人物台词与英文部分保持完全一致。 师姐声音清雅、镇定,语速正常,不故作高冷。 小师妹声音灵动,前半段配合,后半段逐渐出现疑惑和不满,最后恢复平静,带着一点小小的得意。 同一时间只允许一个人物说话。 说话人物口型准确;非说话人物闭嘴,用真实眼神和微表情回应。 背景音为自然风声、远处流水、布料摩擦、铜镜轻响、人物脚步以及一次轻微的留影石灵光声。 配乐极轻,在视觉揭示和最终角色互换时适当退去。 不要罐头笑声,不生成画面字幕或可读文字。 【人物、道具及环境连续性】 师姐、小师妹始终保持图一和图二对应的人脸身份、发型、身材、服装和武器。 全片只有一面铜镜和一枚留影石。 铜镜的实际位置、镜面方向、人物握持方法和反射光线必须相互匹配。 小师妹举镜动作必须符合正常人体活动范围。 师姐的银剑与小师妹的深色钢剑不能交换。 留影石拍下的图像必须与之前展示的实际取景构图一致。 同一地点、地形、建筑、树木、水体、主光方向、天气和空间层次保持连续。 【Seedance 2.5执行要求】 严格30秒,16:9横屏。 优先一次生成完整故事,按照0–4秒、4–10秒、10–18秒、18–25秒、25–30秒五个段落组织。 充分利用 Seedance 2.5 的多人物动作一致性、多参考素材控制、连续镜头叙事与原生音视频生成能力。 镜头可以根据故事需要合理切换,但所有动作和道具状态必须连续。 不加入多余人物、台词、特效、剧情和新地点。 真正的笑点必须清楚呈现在画面里。 最后以人物位置互换完成视觉回收,既构成完整结局,也保留两名女主角未来继续发展关系喜剧的空间。 【Negative Prompt】 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
檢視詳情
0:30
ByteDanceSeedance 2.5
RECRAFT AI — THE WAY YOU IMAGINE Premium 2D animated technology advertisement | Flat graphic illustration | Soft painterly gouache texture | Cinematic visual storytelling CREATIVE INTENT Create an original, visually engaging 2D animated advertisement for Recraft AI. The story follows a young creative professional struggling to choose between different AI generation platforms. His computer screen is filled with browser tabs, generation interfaces, unfinished artwork and confusing workflows. Frustrated, he discovers Recraft AI. He opens the platform, writes a creative prompt and generates an extraordinary piece of artwork. His imagination comes alive. He publishes the creation on X, and the artwork receives attention as likes, reposts, comments and notifications burst out of his computer screen in a spectacular animated sequence. The advertisement ends with the official Recraft logo and tagline: “RECRAFT IS THE WAY YOU IMAGINE.” VISUAL STYLE Sophisticated 2D flat graphic animation with soft painterly gouache illustration. Visual characteristics: * Flat geometric shapes and simplified silhouettes. * Clean, expressive character designs. * Layered cutout environments. * Hand-painted gouache textures. * Subtle watercolor pigment variation. * Visible paper grain. * Imperfect brush edges. * Simplified architectural details. * Controlled graphic shadows. * Rich but restrained color blocking. Use a premium contemporary editorial illustration aesthetic. The animation should feel like an internationally produced technology brand campaign, not a generic explainer video. STRICTLY 2D. No 3D CGI, no photorealism, no plastic-looking characters, no Pixar aesthetic and no anime. CHARACTER DESIGN One young creative professional, approximately 25–30 years old. Contemporary hairstyle, expressive face, modern casual clothing and recognizable body proportions. Design the character using clean geometric shapes, simplified facial features and subtle hand-drawn details. Maintain consistent character design, clothing, facial proportions and colors throughout the advertisement. Facial expressions should communicate confusion, curiosity, concentration, excitement and satisfaction. Movement should feel fluid and natural, with restrained squash-and-stretch animation. ENVIRONMENT DESIGN A modern creative studio illustrated through layered 2D compositions. Large desktop monitor, keyboard, mouse, desk, chair, shelves and minimal accessories. Use simplified geometric architecture and carefully arranged graphic objects. Opening environment: muted charcoal, grey and dusty blue. As Recraft enters the story, introduce brighter colors, warmer highlights and more expressive compositions. Maintain consistent spatial relationships throughout. STORY SEQUENCE SCENE 01 — TOO MANY CHOICES A wide 2D illustrated view of the creative professional sitting at his workstation. His monitor displays multiple overlapping browser tabs and different AI generation interfaces. Small floating windows appear around the monitor, each containing unfinished artwork, loading indicators and confusing interface elements. The character switches between tabs, scratches his head and leans back with a frustrated expression. Camera: Slow lateral 2D camera pan. Animation: Floating windows slide across the screen, browser tabs multiply and interface elements move independently. Visual mood: Slightly chaotic, muted and visually crowded. SCENE 02 — THE DISCOVERY The character notices Recraft AI on his computer. A clean Recraft interface appears on the monitor. The surrounding floating windows gradually disappear, leaving a visually organized workspace. The character leans forward with curiosity. Camera: Controlled push-in toward the monitor. Animation: Interface panels slide into position, cluttered shapes dissolve and the composition becomes cleaner. The official Recraft logo must match the supplied reference exactly. SCENE 03 — THE PROMPT Extreme close-up of the character typing on the keyboard. A prompt appears inside the Recraft workspace. The cursor moves, the character clicks generate and a subtle animated loading sequence begins. The screen gradually fills with an extraordinary generated artwork. Camera: Close-up transitions into a centered monitor composition. Animation: Typography appears through clean graphic motion, interface elements respond naturally and the generated artwork reveals itself through layered illustrated shapes. SCENE 04 — IMAGINATION COMES ALIVE The generated artwork expands beyond the boundaries of the monitor. Its illustrated elements separate into multiple flat graphic layers. Objects, shapes, colors and textures move outward into the studio. The room transforms into an imaginative illustrated environment inspired by the generated creation. Use parallax, overlapping cutout layers, graphic transitions and controlled perspective to create an illusion of depth. Camera: Smooth 2D push-in through the artwork. Animation: Layered shapes expand, rotate, unfold and transition into the surrounding environment. This is the visual climax of the advertisement. SCENE 05 — THE WORLD REACTS The character publishes the finished artwork on X. Show a recognizable social post interface with the artwork. Suddenly, animated engagement elements emerge from the monitor. Heart icons, repost symbols, comment bubbles and notification shapes burst into the studio. They travel across the frame in multiple layers, creating a spectacular three-dimensional impression through 2D perspective. The symbols rotate, bounce, overlap and move toward the foreground. The character watches the engagement with a natural, satisfied smile. Animation: Fluid motion paths, controlled overshoot, squash-and-stretch, layered parallax and expressive graphic transitions. Important: All social-media elements must originate from
檢視詳情
0:13
ByteDanceSeedance 2.5
Create a cinematic 2D anime-style martial arts training sequence featuring a young athletic woman and a tall athletic man sparring inside a traditional martial arts dojo. Characters: - Woman: Young adult, athletic build, black hair tied back in a ponytail, wearing a fitted black sleeveless martial arts training outfit. - Man: Young adult, tall muscular build, short black hair, wearing a dark grey traditional martial arts uniform. Scene and action: Start with a medium-wide cinematic shot of both fighters facing each other in defensive martial arts stances. The woman raises her fists while the man slowly approaches with a focused expression. They begin an intense hand-to-hand sparring sequence. The woman throws quick punches, the man blocks and counters with controlled movements. She dodges his attacks, steps sideways and maintains her fighting stance. The man suddenly launches a powerful jumping kick. The woman reacts quickly, ducks and evades the attack. Continue with dynamic martial arts movements, fast punches, defensive blocks, spinning kicks and agile footwork. Finish with both fighters returning to their fighting stances, facing each other with intense concentration, ready for another exchange. Visual style: Premium Japanese anime animation, cinematic 2D hand-drawn aesthetic, detailed character designs, smooth fluid movement, expressive facial animation, dramatic action choreography, realistic martial arts physics, beautifully animated clothing and hair. Environment: Traditional indoor martial arts training hall, wooden floors, large windows, warm golden sunlight streaming through the windows, atmospheric shadows, subtle dust particles floating in the sunlight. Camera: Cinematic medium shots, dynamic side angles, occasional low-angle action shots, smooth camera tracking, carefully timed action cuts, no unnecessary camera shake. Lighting: Warm golden-hour lighting, dramatic cinematic shadows, beautiful amber highlights, high contrast. Quality: Ultra-detailed anime film quality, fluid 24fps animation, consistent character appearance, professional action choreography, widescreen 16:9, approximately 13 seconds.
檢視詳情
0:30
ByteDanceSeedance 2.5
Create a 30-second ultra-realistic live-action video featuring a naturally beautiful young Korean woman in her early 20s spending time with a friend in an older residential neighborhood in Korea on a warm late-summer afternoon. The video should feel like genuine footage casually recorded on an early 2000s consumer digital camcorder. It should not look like a commercial, K-drama, music video, staged production, or cinematic film. Keep everything simple, candid, engaging, and natural. **Character Consistency** The main character has shoulder-length dark brown-black hair with soft, natural waves, subtle makeup, realistic skin texture, and a relaxed, expressive face. She wears: - A fitted white short-sleeve top. - Light blue straight-leg jeans. - White sneakers. - A small dark brown shoulder bag. - A delicate silver necklace. Maintain perfect character consistency throughout the entire video. Her facial features, hairstyle, clothing, accessories, body proportions, and the same bottled drink must remain identical in every shot. **Location and Environment** A quiet, older Korean residential neighborhood featuring: - Narrow concrete alleys and small side streets. - Low-rise brick villa-style apartment buildings (빌라) and small apartment complexes. - Brick and concrete walls, with potted plants and small rooftop or doorstep gardens. - Parked bicycles. - Utility poles with tangled overhead electrical and cable wires. - A small neighborhood mini-mart or convenience store with an open front and a drink refrigerator. - Shaded walkways, narrow stairways, and apartment entrances with metal gates. The environment should feel lived-in, ordinary, and authentic. Avoid recognizable landmarks, identifiable brands, readable signage, advertisements, or commercial activity. Any signs or labels should be blurred, generic, or out of focus. **Visual Style and Camera Direction** Authentic handheld early 2000s consumer DV camcorder footage. Use: - Natural handheld camera shake. - Imperfect framing and spontaneous camera movements. - Occasional autofocus breathing and focus hunting. - Slight automatic exposure fluctuations. - Soft digital image detail. - Faded, slightly washed-out colors. - Subtle digital video noise. - Natural motion blur. - Occasional accidental zooming. The person filming should behave like a friend casually recording everyday moments. The characters should react naturally to the camera. Absolutely no gimbal shots, drone footage, stabilized camera movements, slow motion, beauty filters, or polished cinematic aesthetics. **Storyboard and Timeline** **00:00–00:05, Leaving the Apartment** She walks out of her apartment building through a metal gate, adjusts her shoulder bag, notices the camera, and smiles naturally. Her friend casually follows behind while filming. **00:05–00:10, Buying a Cold Drink** She buys a bottled cold drink from a small neighborhood mini-mart. She gently presses the cold bottle against her cheek, laughs as the camera zooms in on her face, and casually says in Korean: "오늘 진짜 덥다." (It's so hot today.) **00:10–00:15, Walking Through the Neighborhood** The two friends walk along a quieter residential alley. A gentle breeze moves her hair. She turns back toward the camera, looks directly into the lens, and smiles naturally. **00:15–00:20, A Playful Moment** She sits on a low wall beside an apartment garden. Her friend notices a tiny leaf caught in her hair and points it out. She removes the leaf herself and bursts into natural laughter. **00:20–00:25, Sharing a Drink** They continue walking together. She briefly takes the bottled drink from her friend, takes a casual sip, hands it back, and smiles toward the camera. **00:25–00:30, Saying Goodbye** At the apartment entrance, she turns toward the camera, waves casually, and says in Korean: "내일 봐." (See you tomorrow.) She walks inside the apartment building. Her friend continues recording for one additional second as the door closes. The footage then abruptly cuts to black. **Audio and Sound Design** Use only authentic environmental sounds recorded naturally within the scene: - Footsteps on concrete. - Distant passing traffic. - Bicycle sounds. - Summer cicadas (매미) and other insects. - Ambient residential neighborhood sounds. - Mini-mart refrigerator humming. - Bottle handling and movement. - Rustling leaves. - Faint conversations in Korean and distant neighborhood chatter. Include only natural character dialogue. No background music, voiceover, artificial sound effects, subtitles, captions, explanatory text, logos, watermarks, or random text. **Realism and Physical Accuracy** Every movement must feel spontaneous and physically believable. Ensure realistic: - Walking and sitting movements. - Drinking and handling the bottle. - Hair movement in the breeze. - Facial expressions and hand gestures. - Natural interactions between friends. Avoid exaggerated acting, identity drift, wardrobe changes, distorted hands, extra fingers, duplicate characters, teleportation, CGI-like appearances, or physically impossible movements. **Final Direction:** The finished video should look like a genuine personal memory captured by a friend on an inexpensive early 2000s digital camcorder in Korea. It should feel imperfect, nostalgic, intimate, spontaneous, and completely believable, as if someone found an old home video from the summer of 2003.
檢視詳情
0:30
ByteDanceSeedance 2.5
Create a single continuous self-filmed vlog recorded on a handheld action camera in 16:9 aspect ratio. The camera is always held by the anthropomorphic fox, creating a natural first-person selfie perspective with light wind-induced shake, occasional soft lens flares from sunlight, realistic focus pulls, and dynamic walking/leaning motion. No cuts, transitions, or cinematic effects. Include a subtle realistic digital timestamp in the bottom-right corner that progresses naturally throughout the video. Visual Style: Bright modern 4K digital look with vibrant yet soft natural colors, warm mid-afternoon golden sunlight mixed with gentle shadows, soft bokeh on distant flowers, subtle pollen or floating seeds in the air, crisp details on fur and petals. Slight handheld energy but smooth enough for casual vlogging. Widescreen 16:9 framing. Character: Ember, a 24-year-old anthropomorphic red fox (bipedal, human-like proportions) with soft fluffy reddish-orange fur, white chest and muzzle, bright amber eyes, pointed ears, and a long bushy tail that sways naturally. He wears a simple light denim jacket with rolled sleeves over a soft cream t-shirt and comfortable cargo shorts. Personality: curious, warm, slightly playful and soft-spoken. Location: Starts in a wide open golden meadow filled with tall wildflowers (yellow buttercups, purple lupines, white daisies) surrounded by distant soft green hills and a few scattered oak trees under a clear blue sky with a few wispy clouds. Gentle breeze moving the grass and flowers. Sequence (exactly 30 seconds, continuous take with timestamps): 0:00 – 0:05 Ember lifts the camera, smiles softly at the lens with his ears perking up, and says in a warm, slightly raspy voice: “Hey friends… look at this meadow I just wandered into.” 0:05 – 0:13 He walks slowly through the tall flowers, lowering the camera slightly to show the colorful blooms and butterflies fluttering past while continuing: “The whole field is glowing… and there are so many little butterflies.” 0:13 – 0:20 He crouches carefully among the flowers, gently reaches one paw toward a yellow bloom without picking it, and watches a butterfly land nearby, laughing lightly: “Hi little one… you’re so delicate.” 0:20 – 0:26 He turns the camera to capture both his face and the meadow, then sits cross-legged in a soft patch of grass, letting his tail curl beside him. He gazes around for a few seconds and says quietly: “Places like this make the world feel endless and calm.” 0:26 – 0:30 He stands up, waves one paw at the camera with a gentle smile, and says: “Thanks for coming along today. See you next time.” The camera naturally lowers a little as he continues walking deeper into the meadow.
檢視詳情
0:30
ByteDanceSeedance 2.5
Main Character: A young Korean woman in her early 20s, natural everyday appearance, wearing a faded charcoal-grey sleeveless crop top, loose high-waisted light-wash jeans, black canvas sneakers, and a black cord necklace. Black wavy hair in a messy side ponytail with wispy bangs. Minimal makeup, realistic skin texture, warm personality. Maintain identical face, outfit, hairstyle, and body proportions throughout. Location: A quiet Korean residential neighborhood in late spring. Narrow concrete alleys, modest houses, metal gates, potted plants, bicycles, rooftop terraces, laundry lines, utility poles, overhead wires, and leafy trees. Authentic lived-in surroundings, no shops, cafés, advertisements, or crowds. Visual Style: Extremely photorealistic documentary footage resembling a forgotten personal MiniDV camcorder recording from 2003. Natural behavior, imperfect framing, realistic movement, genuine environmental details, slightly faded colors, soft contrast, mild digital noise and compression artifacts. Camera: Casual handheld recording by a friend. Noticeable camera shake, occasional crooked framing, autofocus hunting, exposure fluctuations, slight motion blur, accidental zooming, and imperfect camera reactions. Absolutely no cinematic movements, stabilization, artificial beauty filters, or modern color grading. SCENE TIMELINE: 00:00–00:05: Outside her house, she waters several flowerpots with a small plastic watering can. Some water accidentally splashes onto her sneakers, making her laugh. The camera struggles to focus. 00:05–00:10: Walking through a narrow alley, she notices an old bicycle leaning against a wall. She rings its bell once, smiles playfully toward the camera, and continues walking. 00:10–00:15: She stops beside flowers growing along a concrete wall, notices a small butterfly, and watches it curiously. A brief breeze moves her hair as the camera struggles to capture the butterfly. 00:15–00:20: Someone calls her from behind a nearby house gate. She turns, smiles, exchanges a casual Korean greeting, and sits on the front steps to eat a small snack. She notices the camera and playfully covers its lens. 00:20–00:25: The camera follows her walking beneath leafy trees along a quiet residential road. Sunlight creates moving shadows on the pavement. She briefly turns toward the camera with a natural smile before looking ahead. 00:25–00:30: At a small residential intersection, she notices a cat sitting beside a wall. She smiles, looks back at the camera, and gives a small wave. The operator accidentally zooms in too far, autofocus loses her face briefly, and the footage abruptly cuts to black. AUDIO: Natural ambient sound only: birds, wind, rustling leaves, distant motorcycles, footsteps, bicycle bell, faint neighborhood conversations, natural Korean speech, and subtle camcorder handling noise. No music, narration, or cinematic sound effects. STRICT REALISM: Prioritize believable human movement, natural expressions, consistent character identity, realistic hands, skin pores, fabric physics, and authentic lighting. Avoid plastic skin, exaggerated expressions, robotic movements, distorted anatomy, morphing, flickering, and artificial-looking backgrounds. FINAL GOAL: A genuinely believable, warm, imperfect early-2000s Korean neighborhood home video, as if casually recorded by a friend during an ordinary morning. Every scene should feel spontaneous rather than staged or AI-generated.
檢視詳情
0:15
ByteDanceSeedance 2.5
Create a 15-second ultra-realistic vertical smartphone vlog featuring the same clearly adult Korean female idol. Make this a completely different concept from the previous fan-meeting vlog: she unexpectedly discovers a small good-luck surprise from her staff just before going on stage. The footage must feel like genuine smartphone footage recorded by the idol herself, with natural handheld movement, realistic front-camera perspective, subtle autofocus shifts, imperfect framing, authentic skin texture, natural facial expressions, realistic hands and fingers, and believable backstage ambience. Avoid polished commercial-video aesthetics. She walks down a backstage hallway filming herself. She suddenly notices something off-camera, turns toward it with a surprised smile, and says softly, “Wait… what?” She enters a small preparation room and discovers a tiny decorated cake and a handwritten “Good Luck!” note waiting for her. She laughs genuinely, picks up the note, and briefly shows it to the camera. She looks back at the camera with a touched smile, takes a tiny bite of the cake, then laughs because a little frosting gets near her lip. A staff member jokingly calls her name from outside the room. She quickly walks toward the stage entrance while filming. The camera catches colorful stage lights and faint audience sounds ahead. She turns the phone toward herself, takes a deep breath, makes a small finger heart, smiles confidently, and says, “Okay, let’s go!” She turns the camera forward and walks onto the stage, ending naturally. Realism requirements: continuous character identity, consistent outfit and hairstyle, physically correct anatomy, no extra fingers, no duplicated objects, no warped hands, no sudden outfit changes, no floating props, no unnatural facial movement, no excessive camera shake, no artificial transitions. Keep the performance spontaneous, warm, and believable, like an actual Korean idol’s personal backstage vlog. Output: 15 seconds, vertical, realistic smartphone video, natural audio ambience, authentic handheld selfie footage.
檢視詳情
瀏覽全部 Seedance 2.5 提示詞探索AI 影片提示詞