身穿立領上衣、斜背黑色小包的韓國偶像系 vlogger，在不同場景分享她的一天

提示詞

[參考角色]
@Image 1 是一個由 13 個分鏡面板組成的 storyboard 網格（第 1 列：4 個面板 | 第 2 列：5 個面板 | 第 3 列：4 個面板），閱讀順序為由左至右、先讀最上列。它定義三件事：(a) 女子的臉部身分、頭髮與肌膚；(b) 她的服裝；(c) 每個鏡頭的場景、構圖與光線。請勿重現網格版面、面板邊框、分割畫面或拼貼接縫——輸出應是一部連續剪輯的影片，任何時刻都只顯示一個全畫面的影像。不要將網格中的任何背景人物用作反覆出現的角色。

[主體設定：Vlogger]
全程都是與 @Image 1 相同的女子。20 出頭、帶有韓國偶像氣質的女子，長直黑髮，具有柔和自然的蓬鬆感，白皙肌膚，大而明亮且富有表情的眼睛，精緻的 V 字臉，小巧鼻子，自然有光澤的嘴唇，清新淡雅的妝容，溫暖親切的表情。
預設服裝（除 Shot 5 外的所有鏡頭）：象牙色立領長袖上衣，採用盤扣，炭灰色寬褲，細肩帶小型黑色斜背包。
僅限 Shot 5 的服裝：短袖紅色絲質旗袍，帶有金色與粉紅色花卉刺繡、立領，右手腕佩戴細手鍊。
她是同一個連續存在的單一人物——絕不要複製或分割她。

[視覺風格]
真實的智慧型手機旅遊影片。26mm 手機鏡頭，深景深，能看見毛孔的真實肌膚質感，輕微 HDR，中性的 iPhone 色彩科學，手持拍攝並帶有小幅自然晃動。未經剪輯的自然抓拍感：不要電影感調色、不要攝影棚燈光、不要光軌、不要人工 bloom 或發光效果、不要額外加入動態模糊效果。不完美的構圖，偶爾輕微傾斜，低光環境中的真實夜間顆粒。

[鏡頭 — 時間戳是每個事件的時間配置，不代表精確到影格的剪接點]

Shot 1 (0:00–0:01.8) Great Wall。在石階上以低角度自拍，身後是瞭望塔與山脊線，天空過度曝光至接近純白。風拂起她的頭髮，她在調整構圖的過程中自然地綻放開朗笑容。她用韓文說：{드디어 중국에 왔어요...!} 硬切。

Shot 2 (0:01.8–0:02.8) Forbidden City。她正大步穿過朱紅色城門，上方是金色屋瓦，身後是寬闊的石造庭院。她沒有停下，回頭看向鏡頭。陰天日光。快速切換。

Shot 3 (0:02.8–0:03.8) Hutong 食物攤。她把剛蒸好的包子推近鏡頭，揚起眉毛，張大嘴巴，做出誇張而驚喜的反應。蒸氣飄過畫面，裸燈泡的室內光線強烈，白色區域略微過曝。快速切換。

Shot 4 (0:03.8–0:04.6) Temple of Heaven 公園。她坐在老柏樹下的木長椅上，抬起下巴，看向遠離鏡頭的方向。斑駁日光，畫面角落有一小塊鏡頭污痕，環境聲在短暫片刻中變得更清晰。以自然動作切換。

Shot 5 (0:04.6–0:06.1) 中式餐廳，紅色旗袍。她坐在擺有點心蒸籠與小菜的圓桌旁，用筷子把食物送到嘴邊，在咬下食物時笑到雙眼緊閉。身後紅燈籠發光，溫暖的琥珀色實用光源，琥珀色光線下的真實肌膚。她用韓文說：{진짜 맛있어!} 在笑聲結束時切換。

Shot 6 (0:06.1–0:07.1) Xi'an 夜市。她站在擁擠的巷道中，手拿羊肉串，頭頂有紅燈籠與霓虹招牌，臉上是不均勻的暖色與粉色光線，周圍人群模糊。她朝鏡頭露出輕微的抿嘴笑。高 ISO 顆粒。快速切換。

Shot 7 (0:07.1–0:08.1) 夜間高速列車。她坐在窗邊，臉孔在玻璃倒影中重疊，外面的城市燈光拖曳成條紋。她望向窗外，不看鏡頭。玻璃上有輕微的鬼影。快速切換。

Shot 8 (0:08.1–0:09.1) Terracotta Army 坑，Xi'an。從欄杆處由後方拍攝，下方一排排陶俑向遠處延伸，她慢慢向前傾身往下看。昏暗的博物館大廳燈光。快速切換。

Shot 9 (0:09.1–0:10.1) ★ 標誌性鏡頭 — Sichuan opera 茶館，Chengdu。側面特寫。一名戴著彩繪面具的 bianlian 表演者站在她身後的明亮舞台上。面具變換時，她睜大雙眼，嘴巴微微張開；紅色、紫色與藍色舞台光直接落在她的肌膚上。拉焦——她的臉部清晰，舞台呈散焦狀態。僅在肌膚上產生投影，不要 bloom。速度：約 60% 慢動作。柔和切換。

Shot 10 (0:10.1–0:10.9) Leshan Giant Buddha。寬幅低角度鏡頭，她在觀景欄杆旁顯得非常渺小，而巨大的石雕佛像從她上方填滿畫面。輕微桶狀變形，天空略微過曝。俐落切換。

Shot 11 (0:10.9–0:11.9) Jinli 燈籠巷，Chengdu。從後方跟拍她走過狹窄巷道，頭頂是一排排紅燈籠，溫暖高光映襯深沉的暗部陰影。她回頭越過肩膀看了一次，接著繼續前行。以向前移動的動作切換。

Shot 12 (0:11.9–0:13.5) Shanghai 天際線。近距離自拍，接著她伸出手臂，Lujiazui 天際線與河流在她身後展開。風吹起的髮絲掠過鏡頭。她的視線從鏡頭移向高樓，笑容變得柔和。自然的手機夜間曝光。她用韓文說：{이런 야경은 차음 봐요...} 在最後一個字上硬切。

Shot 13 (0:13.5–0:15) Chengdu 熊貓基地。她蹲下來，將一根竹筍伸向一隻熊貓幼崽。幼崽抓住竹筍並靠向她；第二隻幼崽推向她的背部，她失去平衡，爆發出真誠的笑聲。手持拍攝、偏離中心、略微不穩。柔和日光，身後是綠色竹林。她用韓文說：{잠깐! 넘어질 육했어!} 淡出至黑畫面。

[音訊]
全程不得有背景音樂。僅使用隨每個場景變化的現場環境音：<開闊石階上的風聲>、<腳步聲與遠處的人群交談聲>、<滋滋作響聲與攤販噪音>、<鳥鳴與樹葉沙沙聲>、<筷子碰撞聲與餐廳低語聲>、<夜市密集人群聲與攤販叫賣聲>、<低沉的高速列車運轉聲>、<安靜的室內回音>、<戲曲鑼鼓聲與觀眾的驚呼聲>、<戶外風聲>、<石板上的安靜腳步聲>、<河風與遠處交通聲>、<竹葉沙沙聲與熊貓咀嚼聲>。她的聲音是自然年輕韓國女子的說話聲，未經處理，具有手持手機在真實距離錄製的效果。

[維持一致性]
每個鏡頭都要有相同的臉孔、相同的頭髮長度與分線、相同的身體比例。除 Shot 5 外，預設服裝在所有鏡頭中保持不變。黑色斜背包在可見時始終位於同一側肩膀。整部影片中只出現一名女子。任何時刻都不得出現畫面文字、字幕、標誌、浮水印、分割畫面或網格。

[Reference Role]
@Image 1 is a 13-panel storyboard grid (Row 1: 4 panels | Row 2: 5 panels | Row 3: 4 panels), read left to right, top row first. It defines three things: (a) the woman's facial identity, hair, and skin; (b) her wardrobe; (c) each shot's location, framing, and lighting. Do NOT reproduce the grid layout, panel borders, split-screen, or collage seams — the output is one continuous edited video showing a single full-frame image at any moment. Do not use any background person from the grid as a recurring character.

[Subject Profile: the Vlogger]
Same woman as @Image 1 throughout. Korean idol-like woman in her early 20s, long straight black hair with soft natural volume, fair skin, large bright expressive eyes, delicate V-line face, small nose, natural glossy lips, fresh minimal makeup, warm and approachable expression.
Default outfit (all shots except Shot 5): ivory mandarin-collar long-sleeve top with frog-button closures, charcoal wide-leg trousers, small black crossbody bag on a thin strap.
Shot 5 outfit only: short-sleeve red silk qipao with gold-and-pink floral embroidery, mandarin collar, thin bracelet on right wrist.
She is one single continuous person — never duplicate or split her.

[Visual Style]
Authentic smartphone travel footage. 26mm phone lens, deep focus, realistic skin texture with visible pores, mild HDR, neutral iPhone color science, handheld with small natural sway. Unedited candid look: no cinematic grading, no studio lighting, no light trails, no artificial bloom or glow effects, no added motion-blur effects. Imperfect framing, occasional slight tilt, real night grain in low light.

[Shots — timestamps are time budgets per event, not frame-accurate cuts]

Shot 1 (0:00–0:01.8) Great Wall. Low-angle selfie on the stone steps, watchtowers and ridgeline behind her, sky overexposed to near white. Wind lifts her hair, she breaks into a natural open smile mid-frame-adjust. She says in Korean: {드디어 중국에 왔어요...!} Hard cut.

Shot 2 (0:01.8–0:02.8) Forbidden City. Mid-stride through the vermilion gate, golden roof tiles above, wide stone courtyard behind. She turns her head back toward camera without stopping. Overcast daylight. Snap cut.

Shot 3 (0:02.8–0:03.8) Hutong food stall. She pushes a freshly steamed baozi close to the lens, eyebrows raised, mouth open in an exaggerated delighted reaction. Steam drifts across frame, harsh bare-bulb interior light, whites slightly clipped. Snap cut.

Shot 4 (0:03.8–0:04.6) Temple of Heaven park. Seated on a wooden bench under old cypress trees, chin lifted, looking away from camera. Dappled daylight, small lens smudge in the corner, ambient sound opens up for a beat. Cut on natural movement.

Shot 5 (0:04.6–0:06.1) Chinese restaurant, red qipao. Seated at a round table with dim sum baskets and small dishes, chopsticks lifting food to her mouth, eyes squeezing shut as she laughs mid-bite. Red lanterns glowing behind her, warm amber practical light, realistic skin in amber. She says in Korean: {진짜 맛있어!} Cut on the end of the laugh.

Shot 6 (0:06.1–0:07.1) Xi'an night market. Standing in the crowded lane holding a lamb skewer, red lanterns and neon signage overhead, uneven warm and pink light on her face, crowd blurred around her. Small closed-mouth smile toward camera. High ISO grain. Snap cut.

Shot 7 (0:07.1–0:08.1) High-speed train at night. She sits by the window, her face doubled in the glass reflection, city lights streaking past outside. She looks out, not at camera. Slight ghosting on the glass. Snap cut.

Shot 8 (0:08.1–0:09.1) Terracotta Army pit, Xi'an. Shot from behind at the railing, rows of clay warriors receding below, she leans forward slowly to look down. Dim museum hall light. Snap cut.

Shot 9 (0:09.1–0:10.1) ★ SIGNATURE — Sichuan opera teahouse, Chengdu. Profile close-up. A bianlian performer in a painted mask stands on the lit stage behind her. Her eyes widen and her mouth opens slightly as the mask changes; red, violet and blue stage light lands directly on her skin. Rack focus — her face sharp, the stage diffused. Projection on skin only, no bloom. Speed: roughly 60% slow motion. Soft cut.

Shot 10 (0:10.1–0:10.9) Leshan Giant Buddha. Wide low angle, she stands very small at the viewing railing while the enormous carved stone Buddha fills the frame above her. Slight barrel distortion, sky slightly blown out. Sharp cut.

Shot 11 (0:10.9–0:11.9) Jinli lantern alley, Chengdu. Following her from behind down the narrow lane, rows of red lanterns overhead, warm highlights against deep dark shadows. She glances back over her shoulder once, then keeps walking. Cut on forward movement.

Shot 12 (0:11.9–0:13.5) Shanghai skyline. Close selfie, then her arm extends and the Lujiazui skyline and river open up behind her. Wind blows strands of hair across the lens. Her eyes drift off camera toward the towers and her smile softens. Natural night phone exposure. She says in Korean: {이런 야경은 차음 봐요...} Hard cut on the last word.

Shot 13 (0:13.5–0:15) Chengdu panda base. She crouches and holds out a bamboo shoot toward a panda cub. The cub grabs it and leans into her; a second cub pushes against her back and she rocks off balance, bursting into genuine laughter. Handheld, off-centre, slightly unsteady. Soft daylight, green bamboo behind. She says in Korean: {잠깐! 넘어질 육했어!} Fade to black.

[Audio]
No background music at any point. Diegetic ambience only, changing with each location: <wind over open stone steps>, <footsteps and distant crowd chatter>, <sizzling and stall noise>, <birdsong and rustling leaves>, <clinking chopsticks and restaurant murmur>, <dense night-market crowd and vendor calls>, <muffled high-speed train hum>, <hushed indoor echo>, <opera gongs and audience gasps>, <open-air wind>, <quiet footsteps on stone>, <river wind and distant traffic>, <bamboo rustle and panda chewing>. Her voice is a natural young Korean woman's speaking voice, unprocessed, at real handheld-phone recording distance.

[Maintain Consistency]
Same face, same hair length and parting, same body proportions in every shot. Default outfit unchanged across all shots except Shot 5. The black crossbody bag stays on the same shoulder whenever visible. Only one woman appears in the entire video. No on-screen text, no subtitles, no logos, no watermarks, no split-screen or grid at any point.