Live-Action to Cartoon Face Morph
In a live-action café setting, a girl's face morphs back and forth between real and cartoon, in a vertical short of just over 8 seconds. The scene props are written in extreme detail — even the record poster on the background wall gets a mention — and that detail is what keeps the morph stable.
PROMPT · Video Prompt
Create with AI整体风格与设定
咖啡馆实拍+AI面部变形/卡通化混合风格。竖屏构图，自然柔和室内光，窗外绿树虚化，背景墙上有黑胶唱片/海报（可见“RAY CHARLES”字样）。
人物：短波波头黑发亚洲年轻女性，白衬衫（胸前有小马刺绣标），卡其短裤，细项链+耳钉，妆容清淡自然。
道具：木桌上有冰饮（带吸管）、手机、小卡片；左侧有白色LV花纹手提包。
情绪基调：活泼、俏皮、可爱、略带夸张的“拍可爱照片”互动感。 分镜拆解（按时间大致划分）
0-1秒：正面坐姿，双手用力拉开嘴角成超大嘴巴，舌头伸出，眼睛向上翻看，做夸张可爱鬼脸。
1-2秒：双手向两侧张开，恢复正常脸，对着镜头甜甜微笑。
2-3秒：双手压在头顶瞬间变形为大头卡通/Q版风格（超大圆脸、巨大眼睛、鼓腮嘟嘴），表情呆萌。
3-4秒：回到真人脸，但叠加超大卡通眼睛+张大嘴巴的惊讶表情，双手高举过头（像在整理头发或做惊讶动作）。
4-5秒：侧头看向一边，微微嘟嘴/思考表情，手臂自然摆动。
5-6秒：卡通头像，闭眼，嘴角上扬，放松微笑，双手放在腿上。
6-7秒：回到真人头像，双手抬起做“爪子”手势，嘴巴张大吐舌，眼睛眯起大笑，超级兴奋俏皮。
7-8.4秒：身体微侧，一手叉腰，另一手拿着红色卡通大花朵（或类似道具），侧脸对镜头甜笑收尾。 整体节奏快，表情切换迅速，真人卡通变形自然流畅，突出“可爱搞怪”氛围。
整體風格與設定
咖啡館實拍＋AI臉部變形／卡通化混合風格。豎屏構圖，自然柔和室內光，窗外綠樹虛化，背景牆上有黑膠唱片／海報（可見RAY CHARLES字樣）。
人物：短鮑伯頭黑髮亞洲年輕女性，白襯衫（胸前有小馬刺繡標），卡其短褲，細項鍊＋耳釘，妝容清淡自然。
道具：木桌上有冰飲（附吸管）、手機、小卡片；左側有白色LV花紋手提包。
情緒基調：活潑、俏皮、可愛、略帶誇張的拍可愛照片互動感。分鏡拆解（按時間大致劃分）
0-1秒：正面坐姿，雙手用力拉開嘴角成超大嘴巴，舌頭伸出，眼睛向上翻看，做誇張可愛鬼臉。
1-2秒：雙手向兩側張開，恢復正常臉，對著鏡頭甜甜微笑。
2-3秒：雙手壓在頭頂瞬間變形為大頭卡通／Q版風格（超大圓臉、巨大眼睛、鼓腮嘟嘴），表情呆萌。
3-4秒：回到真人臉，但疊加超大卡通眼睛＋張大嘴巴的驚訝表情，雙手高舉過頭（像在整理頭髮或做驚訝動作）。
4-5秒：側頭看向一邊，微微嘟嘴／思考表情，手臂自然擺動。
5-6秒：卡通頭像，閉眼，嘴角上揚，放鬆微笑，雙手放在腿上。
6-7秒：回到真人頭像，雙手抬起做爪子手勢，嘴巴張大吐舌，眼睛瞇起大笑，超級興奮俏皮。
7-8.4秒：身體微側，一手叉腰，另一手拿著紅色卡通大花朵（或類似道具），側臉對鏡頭甜笑收尾。整體節奏快，表情切換迅速，真人卡通變形自然流暢，突出可愛搞怪氛圍。
Overall style and setup
A café live-action shoot blended with AI face-morphing / cartoonization. Portrait composition, natural soft indoor light, blurred green trees visible through the window, a vinyl record / poster on the background wall (with "RAY CHARLES" visible on it).
Character: a young Asian woman with a short black bob, wearing a white shirt (with a small embroidered pony logo on the chest) and khaki shorts, a delicate necklace and stud earrings, light, natural makeup.
Props: on the wooden table, an iced drink (with a straw), a phone, and a small card; a white LV-monogram handbag sits to the left.
Emotional tone: lively, playful, cute, with a slightly exaggerated "taking a cute photo" interactive feel. Shot breakdown (roughly by time)
0-1s: Sitting facing the camera, both hands pull the corners of her mouth into an exaggerated wide grin, tongue sticking out, eyes rolled upward, making an exaggerated, cute goofy face.
1-2s: Both hands spread out to the sides, her face returning to normal as she gives the camera a sweet smile.
2-3s: Both hands pressed on top of her head as she instantly morphs into a big-headed cartoon / chibi style (an oversized round face, huge eyes, puffed cheeks, pouty lips), with a dazed, cute expression.
3-4s: Back to her real face, but overlaid with oversized cartoon eyes and a wide-open, surprised mouth, both hands raised above her head (as if fixing her hair or reacting in surprise).
4-5s: Head tilted to one side, a slight pout / thinking expression, arms swaying naturally.
5-6s: Cartoon head, eyes closed, corners of the mouth lifted in a relaxed smile, hands resting on her lap.
6-7s: Back to her real face, both hands raised in a "claw" gesture, mouth wide open with tongue out, eyes squinting as she laughs, looking super excited and playful.
7-8.4s: Body turned slightly to the side, one hand on her hip, the other holding a big red cartoon flower (or similar prop), ending with a sweet smile toward the camera in profile. Overall pace is fast, with rapid switches between expressions; the transformation between the live-action face and the cartoon style should be smooth and natural, emphasizing a "cute and playful" vibe.
全体のスタイルと設定
カフェでの実写撮影＋AIによる顔の変形／カートゥーン化を組み合わせたスタイル。縦位の構図、自然で柔らかな室内光、窓の外の緑の木々はぼやけている。背景の壁にはレコード盤／ポスターがあり（"RAY CHARLES"の文字が見える）。
人物：ショートボブの黒髪のアジア人の若い女性、白いシャツ（胸元に小さな馬の刺繍ロゴ）、カーキ色のショートパンツ、細いネックレスとピアス、メイクは薄く自然。
小道具：木のテーブルの上にはストロー付きの冷たい飲み物、スマートフォン、小さなカードがある。左側には白いLV柄のハンドバッグがある。
情緒のトーン：活発で、いたずらっぽく、可愛らしく、やや誇張された可愛い写真を撮るインタラクション感。 コンテの内訳（おおよその時系列）
0-1秒：正面に座った姿勢で、両手で口角を思い切り引っ張って超巨大な口にし、舌を出し、目を上に向けて、誇張された可愛らしい変顔をする。
1-2秒：両手を左右に開き、通常の顔に戻り、カメラに向かって甘く微笑む。
2-3秒：両手を頭のてっぺんに押し当てた瞬間、大頭のカートゥーン／デフォルメ風スタイルに変形する（超巨大な丸顔、巨大な目、頬を膨らませて口をすぼめる）、表情はぼんやり可愛い。
3-4秒：実写の顔に戻るが、超巨大なカートゥーンの目と大きく開いた口の驚いた表情が重なり、両手を頭の上に高く上げる（髪を整えているような、あるいは驚いた動作のような仕草）。
4-5秒：顔を横に向けて片側を見つめ、わずかに口をすぼめる／考え込む表情、腕は自然に揺れる。
5-6秒：カートゥーンの頭部、目を閉じ、口角を上げ、リラックスした微笑み、両手は膝の上に置く。
6-7秒：実写の頭部に戻り、両手を上げて猫の手のジェスチャーをし、口を大きく開けて舌を出し、目を細めて大笑いする、超テンション高くいたずらっぽい。
7-8.4秒：体をわずかに横に向け、片手を腰に当て、もう一方の手で赤いカートゥーン調の大きな花（または類似の小道具）を持ち、横顔をカメラに向けて甘く微笑んで締めくくる。 全体のリズムは速く、表情の切り替えは素早く、実写とカートゥーンの変形は自然で滑らか、可愛くて茶目っ気のある雰囲気を際立たせる。
전체 스타일 및 설정
카페 실사 촬영 + AI 얼굴 변형/카툰화 혼합 스타일. 세로 구도, 자연스럽고 부드러운 실내광, 창밖 초록 나무 아웃포커스, 배경 벽에 바이닐 레코드/포스터 (RAY CHARLES 문구가 보임).
인물: 짧은 단발머리에 검은 머리의 젊은 아시아 여성, 흰색 셔츠 (가슴에 작은 폴로 자수 로고), 카키색 반바지, 가는 목걸이 + 스터드 귀걸이, 연하고 자연스러운 메이크업.
소품: 나무 테이블 위에 얼음 음료 (빨대 꽂힘), 휴대폰, 작은 카드; 왼쪽에 흰색 LV 모노그램 핸드백.
감정 톤: 발랄하고 장난스럽고 귀여우며, 살짝 과장된 귀여운 사진 찍기 상호작용 느낌. 컷 분해 (시간순 대략 구분)
0-1초: 정면으로 앉은 자세, 양손으로 입꼬리를 힘껏 당겨 초대형 입 모양을 만듦, 혀를 내밀고, 눈을 위로 치켜뜨며, 과장되고 귀여운 익살스러운 표정을 지음.
1-2초: 양손을 양옆으로 펼치며, 원래 얼굴로 돌아와 카메라를 향해 달콤하게 미소지음.
2-3초: 양손을 정수리에 대는 순간 대두 카툰/Q버전 스타일로 변형됨 (초대형 둥근 얼굴, 거대한 눈, 볼록한 볼, 입술 삐죽), 멍하니 귀여운 표정.
3-4초: 실제 얼굴로 돌아오지만 초대형 카툰 눈 + 크게 벌린 입의 놀란 표정이 겹쳐짐, 양손을 머리 위로 높이 듦 (머리를 정리하거나 놀라는 동작처럼).
4-5초: 고개를 옆으로 기울여 한쪽을 바라봄, 살짝 입술을 삐죽이는/생각하는 표정, 팔이 자연스럽게 흔들림.
5-6초: 카툰 얼굴, 눈을 감음, 입꼬리가 올라감, 편안한 미소, 양손을 다리 위에 놓음.
6-7초: 실제 얼굴로 돌아옴, 양손을 들어 발톱 제스처를 취함, 입을 크게 벌려 혀를 내밀고, 눈을 가늘게 뜨며 크게 웃음, 매우 신나고 장난스러운 모습.
7-8.4초: 몸을 살짝 옆으로 틀고, 한 손은 허리에 얹고, 다른 손은 빨간색 카툰풍 큰 꽃 (또는 유사한 소품)을 들고, 옆얼굴로 카메라를 향해 달콤하게 웃으며 마무리. 전체 리듬은 빠르고, 표정 전환이 신속하며, 실사와 카툰 간의 변형이 자연스럽고 매끄러워, 귀엽고 엉뚱한 분위기를 강조함.
✍️ Editor’s Notes
This prompt's ambition is to make the model flip back and forth between live-action and cartoon style four separate times within 8 seconds, rather than doing one single gradual morph — every single second locks down a specific expression and style state, essentially handing the model a state lookup table and forcing it to keep bouncing back and forth instead of finding its own pattern and doing a one-way cartoonification. The background specifically calls out the exact text ‘RAY CHARLES’ on a vinyl record poster, which is a genuinely clever anchor: using one static text element that must be preserved exactly as-is to test and pin down whether the scene drifts while the style keeps flip-flopping — if the background text can stay locked, that proves the scene consistency hasn't been contaminated by the cartoonization. The final prop changes from ‘holding a phone’ to ‘holding a big red cartoon flower,’ which is deliberately designing a standalone memorable beat for the last frame, so the video doesn't peter out with a weak ending.
Related Works
׋![]()
›