[参考图像锁定 — 最高优先级]

REFERENCE IMAGE LOCK — HIGHEST PRIORITY] Throughout the entire video, use the PROVIDED REFERENCE IMAGE as the exact visual identity source for both Chihuahua characters.REFERENCE LOCK: Preserve the exact apple-head shape, upright pointed ear shape and angles, muzzle length and width, eye shape, size and placement, fur color, texture, body proportions, limb thickness, tail shape, relative height difference between both dogs, clothing style, clothing color, clothing patterns, and clothing fit.Both dogs must maintain identical clothing throughout the entire video. No wardrobe changes, no added or removed accessories, and no hairstyle changes. Any clothing inconsistency in any frame is a complete failure.Both dogs must stand and walk on two legs throughout the video. Fur must be photorealistic, finely detailed, with individually visible hair strands. No cartoon rendering, plastic texture, or video-game modeling https://t.co/bwMtCPfelf must remain natural and consistent with the warm evening living-room environment.[SCENE] A cozy living room at night with warm lighting. A pizza box sits on the coffee table. The wife and husband are sitting on the sofa. The wife picks up a slice of pizza, creating a long, stretchy cheese pull.Photorealistic quality, 8K ultra-HD, warm and appetizing colors.[CHARACTERS — BASED ON REFERENCE IMAGE] 1. Wife:Large, overweight white female Chihuahua. Clearly noticeable body weight and volume. Sitting on the sofa while holding pizza. Expectant and excited expression. Wearing the exact outfit from the reference image. Fluffy, snow-white fur. 2. Husband:Small, slim, tan male Chihuahua. Approximately one-third the wife's body size. Sitting beside her. Hungry, eager expression. Wearing the exact outfit from the reference image. [PROP LOCK — EVERY PROP HAS A CLEAR ROLE] Pizza: One pizza box on the coffee table, steaming from 0–3 seconds. Stretchy Cheese: Pulls from the pizza from 3–12 seconds, stretches longer and tighter, then snaps back into the wife's face. Gray Slipper: On the floor, must remain completely stationary until 14 seconds. [LOGIC CHAIN — STRICT SEQUENCE] Every action must happen in the exact order. Skipping any step is a failure.Wife picks up a pizza slice, creating a visible cheese stretch from 0–3 seconds. Wife raises the pizza higher, stretching the cheese longer from 3–6 seconds. Cheese reaches maximum length and becomes tight like a rubber band from 6–9 seconds. Cheese suddenly snaps and slaps directly into the wife's face from 9–12 seconds. Husband points at the wife's cheese-covered face and laughs loudly, saying "HAHAHA." Wife removes the cheese from her face and angrily stares at the husband for exactly 2 seconds. Her eyebrows tighten, her face turns red, and her anger is clearly established. Wife picks up the slipper. Husband begs, "sorry sorry." Wife delivers a slow, comedic BONK to his head at 80% normal speed. The impact trajectory must be clearly visible without motion blur. Husband's eyes turn into dizzy cartoon circles, he falls down, and stars spin above his head. End on a freeze frame. [18-SECOND SEQUENCE — SINGLE CONTINUOUS SHOT] NO CUTS — STRICT TIMELINE — STATIC CAMERA0–3 SECONDS — PIZZA & EXPECTATION Warmly lit living room at night. A steaming pizza box sits on the coffee table, with golden melted cheese.The wife picks up a pizza slice, creating long cheese strands. She looks excited and expectant.The husband sits beside her, hungrily swallowing and watching the pizza.3–6 SECONDS — CHEESE STRETCH The wife raises the pizza higher. The cheese stretches longer, from approximately 10 cm to 30 cm, resembling a golden waterfall.She proudly wiggles the pizza slightly. The cheese strands glisten under the warm lighting.The cheese extension must be clearly visible and progressively increasing.6–9 SECONDS — MAXIMUM TENSION The cheese reaches maximum length and becomes stretched tight like a rubber band.A subtle squeaky "EEEE" sound is heard.The wife tries to take a bite, making the cheese stretch even tighter.The tension must be clearly visible, creating anticipation that something is about to go wrong.9–12 SECONDS — CHEESE SLAP COMEDY IMPACT BEAT.The cheese suddenly snaps and rapidly springs back, hitting the wife's face directly.The cheese covers her nose and mouth like a ridiculous cheese mask.She freezes in place, still holding the pizza slice in the air.The snap-back motion and face impact must be clearly visible.12–14 SECONDS — LAUGHTER & ANGER The husband points directly at the wife's cheese-covered face and laughs loudly:"HAHAHA!"He holds his belly and rocks back and forth while laughing.The wife lifts her paw and pulls the cheese off her face. Cheese strands remain attached to her paw.She disgustedly shakes the cheese off, then slowly turns her head toward the husband.She stares angrily at him for exactly 2 seconds.Her eyebrows tighten. Her face turns exaggerated cartoon red. Steam-like anger puffs appear from her nose.Her anger must be clearly established before she takes any action.14–16 SECONDS — SLIPPER THREAT The wife throws the pizza back into the box.She bends down and picks up the gray slipper from the floor.She raises it above her head, looking furious.The husband instantly stops laughing. His eyes widen in fear.He raises both paws in surrender and quickly says:"sorry sorry"He backs away nervously.16–18 SECONDS — SLOW-MOTION BONK 80% NORMAL SPEED — 20% SLOWERThe wife swings the slipper downward at 80% normal speed.The complete swing takes approximately 1 second.The slipper follows a clearly visible curved trajectory with no motion blur or ghosting.After approximately 1 second, the slipper hits the husband's head with a loud comedic:"BONK!"At the exact moment of contact, freeze the image slightly for 0.2 seconds to emphasize the impact.The husband’s eyes transform into exaggerated cartoon dizzy circles.His tongue sticks out as he falls backward onto the sofa.Three yellow cartoon stars spin above his head.The wife stands proudly with her arms crossed.The pizza box remains on the coffee table.END ON A FREEZE FRAME.[SLOW-MOTION IMPACT — MANDATORY] The slipper swing must use 80% normal speed, meaning the action is 20% slower than normal.The swing must be easy to follow:From raising the slipper to striking the husband: approximately 1 second. The complete arc of the slipper must be visible. No motion blur or ghost trails. The impact must be clearly readable. Freeze the contact moment for 0.2 seconds. The slow motion must not make the hit feel weak. The BONK sound must remain loud and comedic. The action must feel like classic Tom & Jerry slapstick comedy, not realistic violence. [ERROR-PREVENTION & CONTINUITY LOCK] Camera remains locked on a tripod throughout the entire video. No camera movement, shaking, panning, tilting, or zooming. Both dogs remain fully visible in the frame from beginning to end. No characters leave the frame or teleport. Pizza must remain continuously trackable from pickup to return to the box. Cheese must stretch continuously and follow believable elastic physics. Cheese must not suddenly appear on the wife's face without a visible snap-back. Cheese must remain visible on the wife's face from 9 seconds until she removes it. Gray slipper must remain stationary on the floor until 14 seconds. After pickup, the slipper must remain clearly visible and consistent in shape. Fur must not flicker, intersect, or deform. Ear shapes must remain consistent throughout. Lighting must remain stable without sudden brightness changes or flickering. Coffee table and pizza box positions must remain fixed. No background object drifting or disappearing. [KEY RULES — PREVENT LOGIC BREAKS] Clothing must match the reference image 100% throughout the video. Cheese stretching and tightening from 3–9 seconds must be clearly visible. Cheese snap-back and face impact from 9–12 seconds must be clearly visible. Husband must point directly at the wife's cheese-covered face while laughing. Wife must stare angrily at the husband for exactly 2 seconds before picking up the slipper. The slipper impact must be exaggerated, funny, and cartoonish, like Tom & Jerry. No realistic violence, injury, blood, or pain. Husband must finish saying "sorry sorry" before the slipper makes contact. Cheese must only cover the wife's face and must not injure her. Slipper must move at 80% normal speed, never flash by too quickly. Single continuous static shot with no cuts, jumps, or skipped frames. [DIRECTOR'S DETAIL NOTES] Facial Expression Timeline 0–3 seconds: Wife looks excited and expectant. Husband looks hungry. 3–6 seconds: Wife looks proud and pleased with the cheese stretch. 6–9 seconds: Cheese tightens while the wife remains unaware of the impending mishap. 9–12 seconds: Wife freezes after the cheese slap. Husband bursts into laughter. 12–14 seconds: Wife removes the cheese, then develops intense exaggerated anger. 14–18 seconds: Husband begs for forgiveness, followed by a controlled slow-motion comedic impact. Continuity Notes Cheese naturally stretches from 10 cm to 30 cm. Cheese strands become thinner as they tighten. Snap-back must follow believable elastic physics. Cheese remains fixed on the wife's face after impact. After removal, cheese remains attached to her paw until she shakes it off. No sudden disappearance or repositioning of cheese. [SOUND DESIGN — LAYERED AUDIO] Include the following sound effects:Pizza pickup sound. Cheese stretching: "SZZZZ." Cheese tension: subtle "SQUEAK." Cheese snapping: "POP!" Snap-back: "WHOOSH!" Face impact: "SPLAT!" Wife's muffled reaction. Husband's loud laughter: "HAHAHA!" Cheese peeling sound. Cheese shaking sound. Angry grunting. Pizza returning to the box. Slipper pickup sound. Husband begging: "sorry sorry." Slow-motion slipper whoosh, slightly lower in pitch. Loud comedic "BONK!" Spinning stars: magical twinkle sound. [LIGHTING & CAMERA DIRECTION] Warm evening living-room lighting.The cheese strands should appear golden and glossy under the warm light.The cheese snap-back must become the main visual focus.The wife's cheese-covered face should look funny and https://t.co/4AkD7uNN1p the BONK moment:Slight comedic screen shake effect. 0.2-second freeze frame. Camera must remain locked. Spinning stars should have subtle sparkling highlights. [COMPOSITION & FRAMING] Sofa centered in the frame. Coffee table positioned in the foreground. Wife positioned on the left side holding the pizza. Husband positioned on the right side. Cheese stretch centered in the composition. Golden cheese strands form a waterfall-like visual. Final composition: Wife standing on the left, husband lying on the right side of the sofa, pizza box centered on the coffee table. Balanced composition throughout. [ELASTIC CHEESE COMEDY PHYSICS] The cheese tension sequence from 6–9 seconds is the key visual comedy moment.The golden cheese stretches thinner and tighter, creating a delicious yet suspenseful feeling.The audience should feel that something is about to go wrong.The cheese must snap quickly, creating a strong contrast with the slower BONK sequence.The face impact should feel funny and harmless.The wife's reaction should communicate:"What happened to my pizza?"Her frustration should be more about losing the pizza than being hurt.[CHARACTER PERSONALITY ENHANCEMENT] Wife — The Food Lover Eyes light up when she picks up the pizza. Proud expression during the cheese stretch. After the snap-back, her first reaction is to check whether the pizza is still okay. Her obsession with food makes her adorable and funny. The final BONK represents her comedic "food-lover pride." Husband — The Teasing Trouble-Maker Hungry and eager while watching the pizza. Laughs mischievously when the cheese hits his wife. Clearly points at her face while laughing. Immediately becomes scared when she picks up the slipper. Begs for forgiveness before the impact. Comedy Concept: "When a food lover loses her pizza."[VISUAL STYLE] Ultra-photorealistic dogs combined with exaggerated cartoon slapstick physics.Highly detailed realistic fur. Realistic cheese texture and elasticity. Warm cinematic lighting. Exaggerated facial expressions. Disney-Pixar-inspired comedic emotional exaggeration. No plastic-looking fur. No video-game rendering. No realistic violence. [CAMERA SETTINGS] Static wide-angle camera. 16:9 aspect ratio. Eye-level perspective. 24 FPS. Sofa, pizza, and both dogs remain visible throughout. Locked tripod. No camera movement. No cuts. No zoom. No panning. No tilting. [AUDIO & DIALOGUE] Allowed dialogue only:"HAHAHA" "sorry sorry" Prohibited:Dog barking. Cat sounds. Narration. Voice-over commentary. Include natural living-room ambience and all specified sound effects.[NEGATIVE PROMPT] Wardrobe changes, bulldog breed, missing pizza pickup, missing cheese stretch, invisible cheese, unexplained snap-back, missing tension sequence, husband laughing without pointing, missing angry expression, immediate slipper attack, disappearing cheese, disappearing pizza, slipper deformation, motion blur, overly fast impact, flickering frames, fur flickering, teleportation, body deformation, extra limbs, realistic violence, blood, injury, camera movement, cuts, skipped frames, animation rendering, plastic 3D cartoon appearance, inconsistent lighting, inconsistent clothing, changing ear shapes, disappearing props, unrealistic physics.[FINAL SETTINGS] Duration: 18 seconds Aspect Ratio: 16:9 Frame Rate: 24 FPS Motion: 0.55 CFG: 8 Seed: 99999 Style Strength: High Camera: Static, locked tripod Shot Type: Single continuous shot, no cutsFINAL PRIORITY: Preserve character identity, wardrobe consistency, logical action sequence, cheese physics, comedic timing, and the slow-motion BONK impact.

[参考图像锁定 — 最高优先级] 在整个视频中，使用所提供的参考图像作为两只吉娃娃角色的确切视觉身份来源。参考锁定：保留确切的苹果头形状、直立尖耳形状和角度、口鼻长度和宽度、眼睛形状、大小和位置、毛皮颜色、纹理、身体比例、四肢粗细、尾巴形状、两只狗之间的相对身高差异、服装风格、服装颜色、服装图案和服装合身度。两只狗在整个视频中必须保持完全相同的服装。不允许更换服装、不允许增减配饰、不允许改变发型。任何帧中出现的服装不一致都将视为完全失败。两只狗在整个视频中必须用两条腿站立和行走。毛皮必须是照片写实的、精细细致的、可见单根毛发。不允许卡通渲染、塑料质感或电子游戏建模 https://t.co/bwMtCPfelf 必须保持自然，并与温暖的夜晚客厅环境一致。[场景] 一个舒适的客厅在夜晚，温暖的灯光下。咖啡桌上放着一个披萨盒。妻子和丈夫坐在沙发上。妻子拿起一片披萨，形成长长的、可拉伸的奶酪丝。照片写实画质，8K 超高清，温暖诱人的色彩。[角色 — 基于参考图像] 1. 妻子：体型较大、超重的白色雌性吉娃娃。 明显可察觉的体重和体积。 坐在沙发上拿着披萨。 期待而兴奋的表情。 穿着参考图像中的确切服装。 蓬松、雪白的毛皮。 2. 丈夫：体型小、纤细、黄褐色的雄性吉娃娃。 约为妻子体型的三分之一。 坐在她旁边。 饥饿、急切的表情。 穿着参考图像中的确切服装。 [道具锁定 — 每个道具都有明确的角色] 披萨：咖啡桌上的一个披萨盒，从 0–3 秒冒热气。 可拉伸奶酪：从 3–12 秒从披萨中拉出，拉伸得更长更紧，然后弹回拍到妻子脸上。 灰色拖鞋：在地板上，必须在 14 秒前保持完全静止。 [逻辑链 — 严格顺序] 每个动作必须按确切顺序发生。跳过任何步骤都算失败。妻子从 0–3 秒拿起一片披萨，形成可见的奶酪拉伸。 妻子从 3–6 秒把披萨举得更高，奶酪拉伸得更长。 奶酪从 6–9 秒达到最大长度并像橡皮筋一样绷紧。 奶酪从 9–12 秒突然崩断并直接拍到妻子脸上。 丈夫指着妻子满是奶酪的脸大笑，说 "HAHAHA"。 妻子从脸上取下奶酪，怒视丈夫整整 2 秒。她的眉头紧锁，脸变红，愤怒清晰确立。 妻子拿起拖鞋。丈夫恳求，"sorry sorry"。妻子以正常速度的 80% 缓慢、滑稽地砰击他的头。冲击轨迹必须清晰可见，没有运动模糊。丈夫的眼睛变成眩晕的卡通圆圈，他倒下，星星在他头上旋转。定格画面结束。 [18 秒序列 — 单个连续镜头] 无剪切 — 严格时间线 — 静态摄像机0–3 秒 — 披萨与期待 夜晚温暖照明的客厅。咖啡桌上放着一个冒热气的披萨盒，金色融化的奶酪。妻子拿起一片披萨，形成长长的奶酪丝。她看起来兴奋而期待。丈夫坐在她旁边，饥饿地吞咽并看着披萨。3–6 秒 — 奶酪拉伸 妻子把披萨举得更高。奶酪拉伸得更长，从大约 10 厘米到 30 厘米，像金色的瀑布。她自豪地轻轻摆动披萨。奶酪丝在温暖的灯光下闪闪发光。奶酪的延展必须清晰可见并逐渐增加。6–9 秒 — 最大张力 奶酪达到最大长度并像橡皮筋一样绷紧。听到微弱的吱吱声 "EEEE"。妻子试图咬一口，使奶酪拉伸得更紧。张力必须清晰可见，营造出将要出错的预期感。9–12 秒 — 奶酪拍击 喜剧冲击节拍。奶酪突然崩断并迅速弹回，直接击中妻子的脸。奶酪像可笑的奶酪面具一样覆盖她的鼻子和嘴巴。她保持冻结姿势，仍在空中举着披萨片。弹回动作和脸部冲击必须清晰可见。12–14 秒 — 笑声与愤怒 丈夫直接指着妻子满是奶酪的脸大声笑："HAHAHA！"他抱着肚子笑着前后摇晃。妻子抬起爪子把奶酪从脸上拉下来。奶酪丝仍然附着在她的爪子上。她厌恶地甩掉奶酪，然后慢慢把头转向丈夫。她愤怒地瞪着他整整 2 秒。她的眉头紧锁。她的脸变成夸张的卡通红色。鼻子冒出蒸汽般的愤怒气团。她的愤怒必须在采取任何行动前清晰确立。14–16 秒 — 拖鞋威胁 妻子把披萨扔回盒子里。她弯腰从地板上捡起灰色拖鞋。她把它举过头顶，看起来怒火中烧。丈夫立刻停止大笑。他的眼睛因恐惧而睁大。他举起双爪投降并快速说："sorry sorry"他紧张地后退。16–18 秒 — 慢动作砰击 正常速度的 80% — 慢 20%妻子以正常速度的 80% 向下挥动拖鞋。完整挥动大约需要 1 秒。拖鞋沿清晰可见的曲线轨迹运动，没有运动模糊或重影。大约 1 秒后，拖鞋以响亮的喜剧声音击中丈夫的头："BONK！"在确切接触的瞬间，图像略微冻结 0.2 秒以强调冲击。丈夫的眼睛变成夸张的卡通眩晕圆圈。他的舌头伸出，向后倒在沙发上。三颗黄色卡通星星在他头上旋转。妻子骄傲地双臂交叉站着。披萨盒保持在咖啡桌上。定格画面结束。[慢动作冲击 — 强制要求] 拖鞋挥动必须使用正常速度的 80%，意味着动作比正常慢 20%。挥动必须易于跟随：从举起拖鞋到击中丈夫：大约 1 秒。 拖鞋的完整弧线必须可见。 没有运动模糊或重影。 冲击必须清晰可读。 冻结接触瞬间 0.2 秒。 慢动作不应使击打感觉软弱。 BONK 声音必须保持响亮和喜剧性。 动作必须感觉像经典的汤姆与杰瑞闹剧，而不是真实的暴力。 [错误预防与连续性锁定] 摄像机在整个视频中锁定在三脚架上。 没有摄像机移动、摇晃、平移、倾斜或缩放。 两只狗在整个过程中始终完全可见在画面中。 没有角色离开画面或瞬移。 披萨从拿起到放回盒子必须持续可追踪。 奶酪必须连续拉伸并遵循可信的弹性物理。 奶酪不得在没有可见弹回的情况下突然出现在妻子脸上。 奶酪从 9 秒开始必须保持可见在妻子脸上，直到她取下。 灰色拖鞋必须保持静止在地板上直到 14 秒。 拿起后，拖鞋必须保持清晰可见并形状一致。 毛皮不得闪烁、交叉或变形。 耳朵形状必须保持一致。 照明必须保持稳定，没有突然的亮度变化或闪烁。 咖啡桌和披萨盒位置必须保持固定。 没有背景物体漂移或消失。 [关键规则 — 防止逻辑断裂] 服装必须在整个视频中 100% 匹配参考图像。 奶酪从 3–9 秒的拉伸和收紧必须清晰可见。 奶酪从 9–12 秒的弹回和脸部冲击必须清晰可见。 丈夫必须直接指着妻子满是奶酪的脸大笑。 妻子必须在拿起拖鞋前愤怒地瞪着丈夫整整 2 秒。 拖鞋冲击必须夸张、有趣和卡通化，像汤姆与杰瑞。 没有真实的暴力、伤害、血或疼痛。 丈夫必须在拖鞋接触前说完 "sorry sorry"。 奶酪只能覆盖妻子的脸，不应伤害她。 拖鞋必须以正常速度的 80% 移动，绝不能一闪而过。 单个连续静态镜头，没有剪切、跳跃或跳帧。 [导演细节备注] 面部表情时间线 0–3 秒：妻子看起来兴奋和期待。丈夫看起来饥饿。 3–6 秒：妻子对奶酪拉伸感到自豪和满意。 6–9 秒：奶酪收紧，而妻子仍不知道即将发生的意外。 9–12 秒：妻子在奶酪拍击后冻结。丈夫爆发出笑声。 12–14 秒：妻子取下奶酪，然后发展出强烈夸张的愤怒。 14–18 秒：丈夫请求原谅，然后是受控的慢动作喜剧冲击。 连续性备注 奶酪自然地从 10 厘米拉伸到 30 厘米。 奶酪丝收紧时变得更细。 弹回必须遵循可信的弹性物理。 奶酪在冲击后保持在妻子脸上固定。 取下后，奶酪保持附着在她的爪子上，直到她甩掉。 奶酪不会突然消失或重新定位。 [声音设计 — 分层音频] 包括以下音效：披萨拿起声音。 奶酪拉伸："SZZZZ"。 奶酪张力：微弱的 "SQUEAK"。 奶酪崩断："POP!"。 弹回："WHOOSH!"。 脸部冲击："SPLAT!"。 妻子闷闷的反应。 丈夫的大笑声："HAHAHA!"。 奶酪剥离声。 奶酪甩动声。 愤怒的咕哝声。 披萨放回盒子的声音。 拖鞋拿起声音。 丈夫恳求："sorry sorry"。 慢动作拖鞋呼啸声，音调略低。 响亮的喜剧 "BONK!"。 旋转星星：魔法闪烁声。 [照明与摄像机指导] 温暖的夜晚客厅照明。奶酪丝在温暖的光线下应显得金黄且有光泽。奶酪弹回必须成为主要视觉焦点。妻子满是奶酪的脸应看起来有趣 https://t.co/4AkD7uNN1p BONK 时刻：轻微喜剧屏幕震动效果。 0.2 秒定格画面。 摄像机必须保持锁定。 旋转星星应有微妙的闪烁高光。 [构图与取景] 沙发居中在画面中。 咖啡桌位于前景。 妻子位于左侧拿着披萨。 丈夫位于右侧。 奶酪拉伸居中在构图中。 金色奶酪丝形成瀑布般的视觉效果。 最终构图：妻子站在左侧，丈夫躺在沙发右侧，披萨盒居中在咖啡桌上。 整个过程中构图平衡。 [弹性奶酪喜剧物理] 6–9 秒的奶酪张力序列是关键视觉喜剧时刻。金色奶酪拉伸得更细更紧，营造出美味而悬疑的感觉。观众应感觉将要出错。奶酪必须快速崩断，与较慢的 BONK 序列形成强烈对比。脸部冲击应感觉有趣且无害。妻子的反应应表达："我的披萨怎么了？"她的沮丧应更多是关于失去披萨，而不是受伤。[角色个性增强] 妻子 — 美食爱好者 拿起披萨时眼睛发亮。 奶酪拉伸期间自豪的表情。 弹回后，她的第一反应是检查披萨是否还好。 她对食物的痴迷让她既可爱又有趣。 最后的 BONK 代表她喜剧性的"美食爱好者骄傲"。 丈夫 — 调皮捣蛋者 看着披萨时饥饿而急切。 当奶酪击中妻子时恶作剧地笑。 笑着清楚地指着她的脸。 当她拿起拖鞋时立刻变得害怕。 在冲击前请求原谅。 喜剧概念： "当美食爱好者失去她的披萨时。"[视觉风格] 超照片写实的狗