Hyperpop Club to Serene Dawn

全局设置 * 主体：红发女孩（鲜艳的铜色头发，深邃的绿眼睛）。 * 服装： * 居家（造型 ）：酸绿色超大号T恤，印有黄色笑脸图案，粉色短裤，头发凌乱。 * 街头/办公（造型 2）：超亮色羽绒服（电光蓝带霓虹黄细节），紧身紫红色露脐上衣，带反光条的宽大紫色工装裤，厚重的银色运动鞋。头发扎成两个凌乱的高马尾。 * 俱乐部（造型 3）：金属感漆皮短裙，酸橙色上衣，霓虹彩虹色（红/绿/蓝）短款环保皮草外套，厚重的系带靴，长发披肩，脸上有闪片。 * 环境：折衷主义、堆满物品的单身公寓（墙上贴满海报，床头亮着粉色霓虹灯牌，挂满鲜艳衣服的开放式衣架，黑白棋盘格对比印花的床品，豹纹毯子，贴满贴纸的冰箱），闪烁着隧道灯光的镀铬地铁车厢，无菌般明亮的开放式办公室，带有频闪灯和烟雾的昏暗夜总会。 * 氛围/情绪：动感能量，疯狂而自信的节奏，专注目标，最终以极具攻击性的舞蹈爆发力收尾。 * 音乐：硬朗、快速的 Techno 节拍，在结尾处过渡为深沉、有立体感的俱乐部贝斯（Drop）。 * 色彩逻辑：超写实波普风格。明亮的霓虹色服装和折衷主义公寓与灰色的门廊和白色的办公室形成最大对比。 * 风格：超写实，极具冲击力的剪辑。 * 逻辑规则：保持严格的运动连贯性和三个造型的精确切换，不使用魔法特效——仅依靠真实环境中的动态剪辑（匹配剪辑 / Match Cuts）。 镜头脚本 镜头 1： * 摄影机：特写镜头（ECU - 极特写），微距镜头。 * 动作：在刺耳的闹钟声中，女孩的绿眼睛瞬间睁大，直视镜头。瞳孔中反射出房间里霓虹灯牌模糊的粉色光芒。 * 音效（SFX）：刺耳、响亮的电子闹钟声。 镜头 2： * 摄影机：全景（WS），35mm 手持摄影机剧烈运动。 * 动作：节奏性剪辑（向后仰）——她在黑白棋盘格床单的背景下猛地坐起，掀开厚重的豹纹毯子。周围是充满创意的混乱：墙上的贴纸，散落的杂志，挂着鲜艳夹克的开放式衣架。她一个急促的动作，用力将双脚落在地板上。 * 音效（SFX）：脚跟撞击地板的沉闷声，沉重毯子挥动的声音，吸气声。 镜头 3： * 摄影机：中特写（MCU），50mm 镜头急剧滑动。 * 动作：动作剪辑（Cut on action）——在水槽洗脸。她猛地抬起湿漉漉的脸，水珠四溅。 * 音效（SFX）：响亮的溅水声，水龙头刺耳的吱嘎声。 镜头 4： * 摄影机：插入镜头（Insert shot），85mm 镜头焦点转换。 * 动作：画面中心匹配剪辑（Match cut）——亮粉色牙刷在用力刷牙，白色泡沫与她的嘴唇形成鲜明对比。 * 音效（SFX）：快速、剧烈的刷毛摩擦声。 镜头 5： * 摄影机：从冰箱内部向外看，24mm 广角。 * 动作：外面贴满鲜艳贴纸的冰箱门被猛然拉开。穿着绿色T恤的手伸进去；当她抓起能量饮料和苹果时，冷光照亮了她专注的脸庞，背景是色彩斑斓的厨房。 * 音效（SFX）：压缩机的嗡嗡声，罐子碰撞的叮当声。 镜头 6： * 摄影机：插入镜头，50mm 手持摄影机。 * 动作：节奏性剪辑——一个干脆的动作，鸡蛋在煎锅边缘敲碎。 * 音效（SFX）：蛋壳碎裂的清脆声，热油响亮的嘶嘶声。 镜头 7： * 摄影机：中特写（MCU），50mm 镜头居中推进。 * 动作：她咬了一口吐司，瞬间瞥了一眼手腕上的智能手表，然后猛地起身离开。 * 音效（SFX）：干脆的咀嚼声，手表的滴答声，推开椅子的隆隆声。 镜头 8： * 摄影机：插入镜头，35mm 俯视角度。 * 动作：动作剪辑——用力将鲜艳的霓虹色袜子套在脚上。 * 音效（SFX）：袜子皮筋弹在皮肤上的啪啪声。 镜头 9： * 摄影机：中景（MS），35mm 甩镜头（Whip pan）。 * 动作：擦除过渡（Wipe transition）到匆忙换装——她已换上造型 2。拉上电光蓝羽绒服的拉链，抓起背包，头发已经扎成了两个高马尾。 * 音效（SFX）：响亮的金属拉链声，厚重尼龙面料的摩擦声。 镜头 10： * 摄影机：插入镜头，50mm 俯视角度。 * 动作：匹配剪辑——穿着银色厚底运动鞋的脚用力踏入画面，鞋带被一个极具攻击性的拉扯动作收紧。 * 音效（SFX）：厚重鞋底沉重的踩踏声，鞋带摩擦的吱嘎声。 镜头 11： * 摄影机：全景（WS），24mm 跟随镜头（Tracking shot）。 * 动作：她冲出公寓大门（对比：公寓内部可见温暖的粉色霓虹灯，而外部是昏暗、破旧的灰色门廊）。在灰色墙壁的衬托下，她亮紫色的工装裤和蓝色的羽绒服色彩“燃烧”。她跑下楼梯。 * 音效（SFX）：沉重关门的砰砰声，运动鞋在混凝土上快速跑动的脚步声。 镜头 12： * 摄影机：从中景到特写（MS 变 CU），35mm Snorricam 效果（摄影机固定在她身上）。 * 动作：声音桥接（Sound bridge）。她站在地铁车厢里，紧紧抓住扶手。她的脸庞在画面中心保持不动且专注，而背景（穿着灰色衣服的人，玻璃外的车站灯光）以疯狂的速度掠过和摇晃。 * 音效（SFX）：铁轨的轰鸣声，车厢金属的刮擦声，沉重的呼吸声。 镜头 13： * 摄影机：从插入镜头到中特写，50mm 急推镜头（Crash zoom）。 * 动作：跳跃剪辑（Smash cut）——她刷通行证，办公室的玻璃门滑开。她冲进明亮的开放式办公区，将蓝色羽绒服甩在椅背上（身上只穿着紫红色上衣），跌坐在办公桌前。 * 音效（SFX）：闸机的电子滴答声，推拉门的沙沙声，办公椅的吱嘎声。 镜头 14： * 摄影机：过肩镜头（OTS），35mm 手持摄影机。 * 动作：戴着银色戒指的手指极具攻击性地快速敲击笔记本电脑键盘。屏幕上闪烁着图表和表格。周围是白色的办公桌和穿着浅色衬衫的员工。她是唯一的一抹亮色。 * 音效（SFX）：疯狂如机关枪般的键盘敲击声。 镜头 15： * 摄影机：全景（WS），带有快速推进（Crash zoom）。 * 动作：L型剪辑（L-cut）。她用力合上笔记本电脑的盖子。在撞击的瞬间（Smash cut），场景立刻改变——她双手推开夜总会厚重的黑门。她已换上造型 3（彩虹皮草，金属感短裙）。频闪灯直射镜头，她冲进人群，开始激烈地跳舞，直视镜头。 * 音效（SFX）：笔记本电脑重重合上的啪声，这也是深沉、厚重的俱乐部贝斯落下的第一个重音（Drop）。背景中人群的尖叫声。

Use uploaded image @[character_ref] as the permanent character identity reference and uploaded image @[storyboard_ref] as the exact visual timeline and sequencing guide. Follow storyboard timing, shot transitions, framing rhythm, emotional flow, and scene continuity precisely. Expand every storyboard panel into smooth connected motion while preserving natural pacing and visual readability. MAIN CHARACTER: Same exact young East Asian woman throughout all scenes. Female college student around age 20. Long dark soft wavy layered hair with natural movement and strands occasionally falling across face. Warm light skin tone, dark brown eyes, slim proportions, youthful face, small mole under left eye, natural lips, minimal makeup, friendly expression. Maintain exact identity consistency throughout entire video. No face morphing, hairstyle changes, age changes, outfit changes, or body proportion shifts. OUTFIT: Black spaghetti strap top. Loose white cardigan hanging casually from shoulders. Faded blue jeans. Casual aesthetic campus fashion. Preserve exact clothing proportions and textures through all scenes. PERSONALITY: Playful, cheerful, naturally expressive, spontaneous, comfortable talking to camera, campus-girl energy, charismatic but effortless. Frequently smiles, small laughs, subtle eye expressions, natural head tilts, realistic blinking and soft reactions. VIDEO STYLE: Authentic early-2010s Sony Cyber-shot digicam aesthetic using CCD sensor rendering. Entire footage must feel like old memory-card footage discovered years later. Harsh direct flash, slight overexposure on skin, soft grain, digital compression artifacts, LCD pixel texture, slight chromatic softness, imperfect white balance, tiny autofocus hunting, nostalgic camera defects and realistic point-and-shoot rendering. POV RULE: Entire video remains from digital camera LCD POV exactly like storyboard. Viewer constantly sees glowing LCD screen in dark surroundings. Outside LCD screen fades softly into darkness and blur. Camera UI overlays visible entire video: REC icon focus brackets battery icon timestamp "8.1.2012 3:15 AM" 1/30 F3.4 ISO100 Visible LCD glare reflections and realistic screen texture. Duration exactly 15 seconds. SCENE 1 (0:00–0:02) Library interior. Walking slowly between bookshelves while holding camera facing herself. Handheld movement slightly shaky. Small laugh. Character says softly in Indonesian: "Lagi di perpustakaan jam segini, sepi banget tapi nyaman." Camera autofocus briefly searches face. Bookshelves tightly packed with colorful books. SCENE 2 (0:02–0:04) Stops near manga shelf. Looks sideways and laughs naturally. Hair falls over cheek. Slight Cyber-shot zoom movement. SCENE 3 (0:04–0:06) Campus staircase. Walking downstairs while filming herself. Camera shakes naturally with footsteps. Students moving behind. SCENE 4 (0:06–0:08) Glass campus corridor. Walking backward facing camera. Afternoon sunlight entering windows. Soft lens flare. Character smiles and says: "Jadi bagus banget vibes nya hari ini." SCENE 5 (0:08–0:10) Outdoor campus area. Rotates camera briefly toward scenery before returning to face. Trees and students visible. SCENE 6 (0:08–0:10) Mall bookstore. Walking while holding drink. Direct flash creates bright skin highlights and darker background. SCENE 7 (0:10–0:12) Standing in front of store glass. Camera reflection visible. Adjusts hair and laughs softly. SCENE 8 (0:12–0:13.5) Close-up shot. Waves toward camera. Freeze frame briefly. REC icon flashes. Cut to black. Keywords: nostalgic digicam, cyber-shot LCD vlog, college life memory footage, authentic CCD rendering, handheld documentary aesthetic, candid moments, realistic camera flaws, early-2010s aesthetic nostalgia.

Create a cinematic realistic high-speed drone video over Istanbul at golden hour, using the reference image as the exact location and flight-path guide. The drone starts very low above the sea surface, almost skimming the water, facing the Galata shoreline. It accelerates quickly forward over the rippling sea, following the drawn path from the foreground water toward the city. After the orbit, the camera tilts and opens toward the Golden Horn, the Historic Peninsula skyline, sunset reflections, ferries, rooftops, and glowing Istanbul atmosphere. The red drawn line must not appear in the video; it is only a hidden flight-path reference. Add powerful emotional Turkish cinematic music: deep ney melody, dramatic darbuka rhythm, orchestral strings, subtle bağlama texture, epic rising percussion, and a grand heroic crescendo as the drone reaches Galata Tower. No text, no subtitles, no logos, no graphic overlays.

Create a 15-second ultra-realistic 16:9 landscape lifestyle video using the provided reference image as the exact character reference. Preserve his face, identity, hairstyle, body proportions, skin tone, and overall appearance consistently throughout. Do not change his identity. 0–3 sec — Kitchen: Medium-wide shot. He happily enters his kitchen, cracks eggs into a bowl, and starts preparing breakfast with energetic but natural movements. 3–6 sec — Cooking: Side three-quarter shot. He pours the eggs into a pan and cooks them, smiling casually. Realistic steam, utensils, hand movements, and cooking physics. 6–9 sec — TV: He briefly walks into the adjacent living room, picks up the remote, turns on the TV, glances at the screen with a small smile, then heads back toward the kitchen. 9–12 sec — Finishing breakfast: Back in the kitchen. He checks and stirs/flips the food, looking pleased, then plates the finished breakfast. 12–15 sec — Eating: Front three-quarter shot at the dining table. He sits down, looks happily at his breakfast, smiles, and takes his first bite. Style: Photorealistic casual daily-life vlog, authentic home environment, happy energetic mood, natural acting, realistic human movement, subtle handheld smartphone camera, natural autofocus, believable depth of field, 24fps. Lighting: Soft natural morning daylight mixed with realistic indoor lighting. Maintain: Consistent face and clothing, realistic hands/fingers, accurate object interaction, consistent kitchen/living-room layout

Generate a 15-second horizontal 16:9 original cyberpunk action anime video from the provided first frame. CRITICAL ENTITY LOCK: There must be exactly 2 main characters in the entire video: KAI (male hacker in techwear trench coat) and REN (female cyborg enforcer with mantis blades). Do not add corporate guards, extra droids, secondary mercenaries, or flying SWAT units. Maintain strict visual consistency for Kai's cyan eye, orange gauntlet glow, trench coat, and Ren's black armored suit, crimson heat vents, and mantis blades throughout all cuts. Character identity: KAI: Male hacker, charcoal techwear trench coat, cyan cybernetic eye, neon orange holographic gauntlet, high-tech stealth parkour movement. REN: Female cyborg enforcer, matte-black armored bodysuit, glowing crimson vents, silver ponytail, dual retractable mantis blades, hyper-speed combat. Video style: High-budget Japanese anime feature film quality (90s Ghost in the Shell meets Psycho-Pass sakuga action), rainy Neo-Tokyo noir atmosphere, fluid blade combat physics, high-contrast neon reflections, hyper-kinetic pacing. Camera and pacing: Hyper-fast, relentless action pacing blending high-tech stealth with lethal velocity: 0.0s - 3.0s: High-speed tracking camera as Kai drops onto rainy rooftop and deploys holographic decoys to dodge Ren's glass-shattering skylight drop. 3.0s - 7.5s: Low-angle arc tracking as Ren speed-blitzes decoys while Kai slides under obstacles and swings via grapple wire. 7.5s - 12.0s: Intense close-combat clash; EMP bullet deflection spark frame leading to Kai launching an orange virus ground shockwave. 12.0s - 15.0s: High-speed server data steal followed by Kai leaping backward off the skyscraper edge into rain and neon twilight, ending on a crisp cliffhanger freeze. Action timing: 0.0s - 2.0s: Infiltration start. KAI drops onto wet rooftop. REN shatters the glass skylight, landing with glowing crimson mantis blades. KAI fires 3 holographic orange decoys. 2.0s - 4.5s: REN slices two decoys in red energy arcs. Real KAI slides under heavy steel pipes, firing a grapple line to swing upward toward the server terminal. 4.5s - 7.5s: REN wall-runs up a steel crane, sparks flying. KAI swings into frame, firing an EMP pistol shot. REN deflects the bullet with her blade in a blinding flash. 7.5s - 10.5s: Both hit the rain-soaked helipad. KAI slams his gauntlet down, sending neon orange malware streams locking REN's armored joints in digital glitch artifacts. 10.5s - 13.0s: KAI vaults past REN, snatching the glowing quantum data drive from the rooftop vault. REN purges the virus with a burst of red steam from her armor vents. 13.0s - 15.0s: KAI leaps backward off the rooftop edge into the rainy neon abyss holding the drive. REN looks down from the ledge. Final cinematic freeze frame. Motion quality: Fluid 2D animation, realistic rain physics, precise blade trajectory rendering, flawless holographic glitters, zero body warping, crisp line art consistency. Environment: Rain-slicked Neo-Tokyo rooftop, glowing neon kanji signs, holographic security barriers, torrential night rain, steam vents, wet asphalt reflections. Final output: 15 seconds, horizontal 16:9, original high-budget cyberpunk anime, exactly 2 main characters, explosive velocity, lethal stealth action, no subtitles, no watermarks, no logos.

Seedance 提示词｜互换佩剑后的剑路暴露一、项目定位电影级写实质感，纯古风中国仙侠美学。整体风格融合：克制的关系型喜剧优雅清晰的武侠身体调度古典视觉喜剧节奏 Arri Alexa 电影机质感清晰稳定的面部细节细腻胶片颗粒自然体积光核心反转：两名女子突然发现，原来对方早就偷偷研究过自己的剑法。要求：两个人始终聪明、强大、有体面不能通过谁出丑制造笑点不能通过能力不足制造笑点喜剧来自人物关系、默契暴露与克制反应二、叙事原则剧情只能由两名人物自己的以下因素推动：选择对白眼神剑术动作身体调度停顿与反应节奏环境必须始终是一个鲜活但叙事中立的活世界：水体持续流动山雾按真实地形游走不同纵深植被以不同幅度响应自然风云层与光线缓慢变化反光表面随摄影机角度自然改变远景保持真实视差与空间环境声但环境绝不能：触发剧情解决问题改变人物目标制造笑点强调包袱替人物做出判断三、参考控制 1. 人物身份锚点 @图片 1：严格作为角色 ID A 剑仙师姐身份锚点 @图片 2：严格作为角色 ID B 小师妹身份锚点 2. 环境参考规则本轮新上传的全部背景和地点参考图共同决定同一套环境 DNA。正式构图前，先无声推演：真实地形建筑语言空间尺度材质年代植被生态水系走向天气云层山雾运动主光方向反光关系整体彩色空气纵深人物真实可行的通行动线然后重新规划成：一个本轮独有、完整统一、此前从未出现过的新世界。要求：不机械复制任何单张背景图保留参考世界身份与逻辑所有镜头共享同一地理空间与环境系统四、角色设定角色 ID A｜剑仙师姐同一个 @图片 1 剑仙师姐。 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。始终穿同一套：白色刺绣真丝汉服半透明分层宽袖银色腰封玉佩白色布靴持 唯一一柄银色长剑。角色 ID B｜小师妹同一个 @图片 2 小师妹。 20–25 岁东亚女性，圆润灵动的脸，黑发编辫，身形娇小。始终穿同一套：青绿色亚麻汉服深色腰带木簪黑色布鞋持 唯一一柄深色钢剑。五、核心表演逻辑这段戏的笑点不是“谁输了”，而是：两人都很强两人都很克制两人都嘴硬两人都在偷偷观察、学习对方真相暴露后，仍然不肯直接承认自己在意对方所以人物反应必须是：微妙克制聪明有分寸不夸张不闹剧化六、分镜脚本 0–5s｜全景或远景在根据本轮参考图真实空间逻辑自然规划出的主要练剑区域中，两个人从鞋到头完整面对面站立。同一个白衣剑仙说道： “今日换剑，看你离了自己的剑，还剩几分本事。” 随后两人同时以正式、双手递剑的动作交换彼此长剑。要求：两人动作端庄、讲究礼数剑的交换过程清楚易读观众必须明确看清两柄剑已经互换摄影机稳定、电影化，完整建立空间与人物关系 5–10s｜中景或牛仔景保持同样两名女子、同样服装、已经互换的同样两柄长剑和完全一致的地理空间。同一个小师妹此时手持银色长剑，完成一组干净流畅、明显带有师姐惯用剑路特征的动作。同一个剑仙师姐手持深色钢剑，立刻以一组短促精准、带有小师妹个人特点的反角度剑势回应。两个人同时停住。小师妹缓慢抬眼问： “你怎么会我的剑？” 师姐平静回答： “你又怎么会我的？” 完整停顿半拍。两个人同时意识到真正的问题。要求：双人动作简单清楚、容易识别 \剑路区别要足够明确 \喜剧点来自“彼此都暴露了” \背景继续自然运行，但零剧情因果 10–15s｜特写或极近特写同一个小师妹首先移开视线，故作随意地回答： “偶尔看过。” 同一个剑仙师姐淡淡说道： “我也是。” 两人的视线再次碰到一起。小师妹眯起眼睛： “偷看？” 师姐认真纠正： “观摩。” 极近特写：两个人以完全同步、干净利落的动作，将彼此手中的长剑重新归还给主人，然后分别朝相反方向转身。摄影机捕捉到两个人脸上几乎完全相同、极力克制却明显存在的轻微笑意。停顿一拍后，两个人又在完全同一时刻回头偷看对方，并且在发现对方也回头的一瞬间，同时把笑容收得干干净净。要求：喜剧通过回头时机一致形成两人都不能真正失态大笑结尾要有默契被发现的余味保持克制、优雅、聪明的关系感七、动作设计要求剑术动作必须清楚、利落、容易辨认不要复杂炫技到看不清小师妹用银色长剑时，动作要能看出她在模仿或熟悉师姐的路数师姐用深色钢剑时，动作要能看出她同样熟悉小师妹的习惯两种剑路特征必须可区分人物身体调度交换剑时动作端庄正式出手、停住、抬眼、移开视线、回头偷看都必须精确喜剧节奏依赖停顿、对视、同步动作与微表情绝不靠夸张肢体乱动制造笑点微表情小师妹更灵动，但要克制师姐更平静，但并非没有情绪暴露“偷学”后，两人都要有短暂的不自然最后的笑意必须轻微、真实、几乎同时出现八、摄影与镜头要求 16:9 横屏三个连续清晰镜头 Arri Alexa 电影观感观察式摄影，稳定但有空间呼吸全景先建立空间与角色关系中景重点呈现互换后的剑路暴露特写重点捕捉嘴硬、心虚与默契感前景、中景、远景保持持续自然视差空间关系、人物站位、剑的归属必须始终清楚九、声音与环境要求原生同步普通话对白口型与声音精准同步真实空间环境声风、水、远景自然声持续存在不使用夸张喜剧音效不用环境帮忙打包袱剑术动作可配少量真实金属与衣料声十、技术规格总时长：15 秒画幅：16:9 横屏三个连续清晰镜头原生同步普通话对白不生成字幕始终只出现两名可见角色人物面部、发型、服装、两柄长剑和地理关系全程稳定真丝衣料和头发物理效果真实前景、中景、远景保持持续自然视差背景始终鲜活但在叙事上完全中立整体结构按照 Seedance 2.0 官方多模态音视频参考和 15 秒多镜头生成能力组织十一、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

GLOBAL STYLE NOTES: • lighting_philosophy: High-contrast night storm lighting, strobe-like lightning flashes, practical headlights. • color_grade: Deep Stark Void (black night with stark white/blue lightning flashes). • setting: Muddy dirt road, apocalyptic night storm, bending trees, torrential rain. • ambience: Deafening wind, heavy rain, engine roaring, thunderclaps. • characters: - character_1: visual_anchors: Zoriana, 20-year-old Eastern European woman, wearing a dirty red shiny evening dress, smeared black mascara under both eyes, wet tangled red hair falling over shoulders, hands gripping steering wheel. SEQUENCE LIST: SHOT 1 (0-4s) Close-Up • vfx: atmospheric rain interaction on glass, strobe lightning flashes. Speed: 100%. • camera_motion: locked-off camera with heavy handheld tracking shake. • action_visual: Zoriana aggressively grips steering wheel inside vintage SUV. Windshield wipers swipe frantically clearing thick rain. She breathes heavily with a panicked expression. Vehicle cabin violently shudders from wind physics. • exit: (HARD CUT TO) SHOT 2 (4-9s) Wide Shot • vfx: volumetric rain, flying debris, viscous mud simulation from tires. Speed: 100%. • camera_motion: tracking shot (lateral), flying low to the ground parallel to moving vehicle. • action_visual: Brutal vintage SUV speeds forward on muddy dirt road. Massive mud clumps fly from massive tires into the camera. Deep background features a forming tornado funnel and trees bending horizontally to the ground under heavy wind. • exit: (WHIP PAN TO) SHOT 3 (9-13s) Medium Shot • vfx: rigid body impact (falling tree), momentum conservation on braking. Speed: 100%. • camera_motion: static shot positioned from passenger seat facing the driver. • action_visual: Massive tree crashes onto the road directly ahead. Zoriana slams on brakes. Momentum throws her violently forward against the seatbelt. She opens her mouth screaming in terror and instinctively raises right arm to shield her face. • exit: (MATCH CUT TO) SHOT 4 (13-15s) POV • vfx: electric arc/lightning (point-to-point), high-intensity bloom/glow. Speed: speed ramp (deceleration) right before impact. • camera_motion: push in rapidly toward windshield. • action_visual: Driver's perspective looking out. SUV skids out of control toward the massive fallen tree trunk. Milliseconds before collision, a gigantic blinding lightning bolt strikes the hood directly. Pure white illumination overtakes the entire frame. • exit: (SMASH CUT TO)

2D anime action animation, feature-film quality, bold clean ink outlines with varying line weight, flat cel shading with hard-edged shadow shapes, rich saturated comic-book colouring, dramatic rim lighting, hand-drawn impact frames, radial and directional speed lines on fast movement, animated on twos with snappy pose-to-pose keyframes and brief holds. Non-photorealistic, never 3D render, never CGI. 16:9. 15 seconds, SIX shots, hard cuts only. Motion starts on frame 1, no opening freeze. Nobody speaks, no dialogue, no lip-sync. THE LOCATION: the gutted interior of an abandoned concrete building at night. Broken windows down one wall throwing hard shafts of blue moonlight through drifting dust, exposed rebar, fallen chunks of masonry, shattered glass across the floor, graffiti on the far wall in abstract unreadable marks. Deep blue-black shadow everywhere the moonlight does not reach. THE FIGHTER: a lean young man in his early twenties, close-cropped dark hair, sharp jaw, a single scar through one eyebrow. An olive bomber jacket worn open over a black tank top, dark cargo trousers, worn trainers, a thin gold chain at the throat, hands wrapped in dirty white tape. Identical in every shot. THE OPPONENT: a heavier man in his thirties, broad and thick-set, shaved head, a scarred blunt face. A dark grey hooded jacket with the hood down, black trousers, heavy boots. Identical in every shot. PALETTE LOCK: cold blue-black night, hard white-blue moonlight, grey concrete, olive green. ONE single accent — the warm gold of the chain and a hot yellow used only in impact flashes. Nothing else is warm. CAMERA: hard lateral tracking that moves with the fighters, low angles looking up, and locked-off frames on impacts. The camera never floats free and never drifts slowly. It either tracks hard with a body or holds absolutely still. SHOT 1 — 0.0s to 2.5s — CAMERA: wide, low angle, tracking hard laterally right to left. INTENT: open mid-action so there is no ramp-up to sit through. The fighter sprints through the ruined interior toward the opponent, moonlight shafts strobing across him as he passes each broken window, dust and grit kicking up behind his trainers, directional speed lines raking the frame edges. HARD CUT. SHOT 2 — 2.5s to 5.0s — CAMERA: close, low angle, locked off. INTENT: one clean readable strike so the fight has weight before it gets fast. He plants and throws a straight right. The frame holds on a single hard-drawn impact frame as the punch lands — radial white speed lines bursting outward from the point of contact, a hot yellow flash, the opponent's head snapping sideways. HARD CUT. SHOT 3 — 5.0s to 7.5s — CAMERA: wide, low angle, locked off, shaking hard. INTENT: physical consequence — the environment takes the hit too. The opponent is thrown backward into a concrete pillar, which cracks and sheds a burst of masonry chunks and dust. Shattered glass across the floor jumps and scatters. He drops to one knee. HARD CUT. SHOT 4 — 7.5s to 10.0s — CAMERA: close on the opponent, low angle, locked off. INTENT: the turn — the fight goes both ways. He rises fast and drives a heavy hook upward. Hard impact frame as it lands, radial speed lines and a hot yellow flash, the fighter's feet leaving the ground. HARD CUT. SHOT 5 — 10.0s to 12.5s — CAMERA: tracking hard with the fighter's body as he flies backward. INTENT: sustained motion between two impacts, so it does not become a slideshow of poses. The fighter is hurled backward through the air across the room, the ruined interior tearing past behind him in heavy directional speed lines, and he crashes down through a heap of broken masonry, dust exploding upward around him. HARD CUT. SHOT 6 — 12.5s to 15.0s — CAMERA: low angle, locked off, very slow push-in. INTENT: end on the stand-up, not the knockdown — the fight is not over. Dust settling. The fighter pushes himself up out of the rubble onto one knee, then stands, head lowered, taped fists closing. A shaft of blue moonlight falls across his face, blood at the corner of his mouth, the gold chain catching one glint. He looks up. The frame holds. Six hard cuts at 2.5s, 5.0s, 7.5s, 10.0s and 12.5s. No dissolves, no fades, no slow motion, no speed ramps. Continuous motion within every shot except the deliberate held impact frames. AUDIO: fast footfalls on concrete and glass, cloth snapping, two heavy bass-weighted impacts, concrete cracking, masonry falling, glass scattering, dust settling, ragged breathing at the end. No music, no voice, no dialogue. NEGATIVE: photorealism, live-action, 3D render, CGI, smooth interpolation, motion blur instead of speed lines, soft gradients, airbrush shading, digital painting look, blood spray, gore, dismemberment, weapons, guns, knives, readable text, letters, words, numerals, readable graffiti, on-screen text, captions, subtitles, logos, watermark, changing costume, changing hair, changing face, extra fighters, cloned characters, dissolves, fades, slow motion, speed ramping, opening freeze, static first frame, anyone talking, lip-sync.