用於文字和圖像提示詞的 Veo AI 影片生成器

描述場景或帶入圖像來引導其第一格畫面。在 CanvasSeed 上使用 Veo 進行文字轉影片與圖像轉影片創作,然後在生成前選擇可用的時長、取景與輸出設定。

0/10000
所需點數—

正在載入影片模型……

Veo:從想法到你的下一部影片

相關靈感 · Seedance 2.0Second Beauty Product Commercial (Vertical 9:16)
檢視提示

10-Second Beauty Product Commercial (Vertical 9:16) Style: Ultra-realistic luxury beauty commercial, soft pink aesthetic, premium cosmetic advertising, clean studio lighting, cinematic, high-end skincare/makeup brand, shallow depth of field, smooth transitions, 4K HDR. Scene 1 (0–2s) – Hook Shot: Close-Up | Front Angle | Fast Push-In A beautiful young woman looks directly into the camera with a surprised and curious expression while holding a matte lip cream beside her face. She slightly raises her eyebrows and opens her lips naturally. Dynamic bold text appears: "Lip Cream That Lasts All Day?!" Soft pink background with glowing highlights. Scene 2 (2–4s) – Product Hero Shot: Hero Shot | Slight Low Angle | Smooth Dolly-In The lip cream bottle and premium packaging stand upright on a glossy reflective surface. Elegant light streaks create luxurious reflections across the product. Floating particles shimmer around the bottle. Animated text appears: "NEW COLOR" and "Matte Last Lip Cream." Scene 3 (4–6s) – Texture Demo Shot: Extreme Macro Close-Up | Macro Push-In The applicator slowly glides across soft lips, revealing an ultra-smooth matte finish with rich pigment and detailed texture. The lipstick blends evenly without streaks. A glowing badge appears: "12 HOURS" with subtle sparkle animation.

開啟結果

附帶運鏡簡報的場景描述

使用文字轉影片來一起描述主體、動作、場景和攝影機運動。Veo 提供了一種從書面簡報探索鏡頭的方式。給場景一個清晰的方向,並檢視結果的運動和視覺細節。

相關靈感 · Seedance 2.0Ultra-realistic cinematic scene at a destroyed Gwanghwamun Gate. Cracked
檢視提示

Ultra-realistic cinematic scene at a destroyed Gwanghwamun Gate. Cracked roads, burning vehicles, debris, and smoke fill the square. Damaged statues of Sejong the Great and Yi Sun-sin in background. Handheld camera, shallow depth, motion blur, heavy impacts, sparks, no fantasy effects. [Character] Asian female (exact match to @image_1), torn white dress, covered in dust, calm and focused. [Armor] Modular exosuit (purple + silver-teal from @image_2). Pieces fly in and lock with metallic precision, sparks, and impact sounds. [Enemies] Fast grey-black beasts with claws and glowing red eyes attacking from all sides. Shots (0–15s) 0–2s: Wide shot. Hero runs through ruins as monsters chase. Armor pieces approach fast. 2–6s: Boots, legs, waist, torso attach mid-run with metallic locks. 6–9s: Helmet seals. Purple energy glows. Camera orbits as enemies close in. 9–15s: Combat—punch sends one flying, second thrown into explosion, final hit creates ground shockwave. Hero ends in a half-kneeling pose.

開啟結果

以圖像作為起點

當現有圖片應引導鏡頭時,使用圖像轉影片。描述主體或攝影機應如何在該圖像之外移動。參考資料有助於建立場景,而生成的運動仍需檢視。

相關靈感 · Seedance 2.0POV first-person perspective. The character is an ultra-realistic tiny
檢視提示

POV first-person perspective. The character is an ultra-realistic tiny human, only the size of a human palm. He is **not** a 3D model, cartoon, CGI, toy, figurine, or miniature statue. He looks like a real miniature person with lifelike skin, natural hair, realistic facial features, authentic clothing, and convincing human movements. Everything in the scene is real, including the table, the owner's hands, the ice cream cone, the lighting, and the environment. The tiny man is sitting on a real wooden table. The owner reaches in with a real human hand and gently pokes the tiny man's cheek with a finger. The tiny man immediately looks annoyed, frowns deeply, pushes the giant finger away with both hands, turns his back, puffs out his cheeks, and sulks with crossed arms. The owner then brings a real ice cream cone into the frame and gently taps the top of the tiny man's head with it. The tiny man glances sideways at the ice cream. His eyes instantly light up with excitement. His expression changes from grumpy to delighted. He quickly turns around, eagerly grabs the giant ice cream cone with both hands, happily licks the melting ice cream, leaving a little cream around his lips. The owner gently pats the tiny man's head with a real finger. The tiny man smiles politely with a grateful and satisfied expression, gives a respectful nod, then immediately hugs the giant ice cream cone and runs toward the corner of the real table. He sits down happily and continues eating the ice cream with obvious enjoyment. Hyper-realistic tiny human illusion, miniature real person, macro photography, true-to-life human proportions, natural facial expressions, realistic skin texture with visible pores, lifelike eyes, authentic clothing physics, smooth natural body movements, perfect scale interaction with giant real objects, cinematic realism, physically accurate lighting, natural contact shadows, seamless integration into the real environment, shallow depth of field, handheld macro camera, one continuous shot, no cuts, no dialogue, no subtitles.

開啟結果

你可以檢視的模型設定

選擇已啟用的 Veo 版本,並使用為其顯示的時長、長寬比和解析度選項。不同版本可能提供不同的組合。報價反映當前選擇,以便你在提交前比較選項。

相關靈感 · Seedance 2.0Create a 10-second cinematic travel vlog montage. Use the uploaded woman s
檢視提示

Create a 10-second cinematic travel vlog montage. Use the uploaded woman’s face consistently throughout every shot. She is a stunning young female travel creator with long dark wavy hair with curly bangs,wearing fashionable oversized Japanese streetwear (burgundy long leather jacket,shoulder bag), natural makeup, cheerful personality. The entire video should feel like real smartphone footage, with handheld camera movement, imperfect framing, accidental lens flare, slight motion blur, rolling shutter, high ISO noise at night, autofocus hunting, overexposed highlights, slight barrel distortion from a phone wide-angle lens, tiny lens smudges, and natural camera shake. No polished commercial look. Story Flow (10s) 0.0–0.8s Low-angle selfie in front of Tokyo Tower. She smiles naturally while adjusting the phone. Slightly tilted horizon. Bright overexposed cloudy sky. Tiny fingerprint on lens. 0.8–1.6s Walking through Shibuya Crossing at night. Looks back at camera while walking. Neon reflections on wet road. Strong motion blur. Quick whip-pan transition. 1.6–2.4s Inside a Japanese convenience store. Holding an onigiri close to camera. Playful eyebrow raise cutely . Harsh fluorescent lighting. Background shelves softly blurred. 2.4–3.2s Sitting alone in Yoyogi Park. Quiet candid moment. Looking away from camera. Soft sunlight filtering through trees. Slight focus miss. 3.2–4.0s Eating freshly made takoyaki. Instant reaction after first bite. Warm lantern lighting. Face slightly blurred from movement. Laughing naturally. 4.0–4.8s Standing in front of a glowing vending machine. Blue and pink lighting on face. Pressing drink button casually. Visible night grain. 4.8–5.6s Reflection in Tokyo tram window. Layered city lights reflected on glass. She looks outside thoughtfully. Train moving creates light streaks. 5.6–6.4s Walking beneath a Torii gate. Low-angle tracking shot from behind. Turns her head briefly. Motion blur. Bright halo from sky. 6.4–7.2s Immersive digital art exhibition. Blue, purple and gold lights moving across face. Looking upward in amazement. Natural low-light noise. 7.2–8.0s Wide-angle low-angle shot beside the giant silver bullet sculpture. She appears small beneath the sculpture. Phone lens distortion. Sky slightly blown out. 8.0–8.8s Walking through a lantern-lit alley. Camera follows behind. She turns around with a smile. Warm lantern glow. Natural night grain. 8.8–10.0s Rooftop selfie overlooking Tokyo skyline. Wind blows hair across camera lens. Slightly tilted horizon. Overexposed city lights. She laughs naturally while extending one arm toward the skyline. Editing Style * Fast whip-pan transitions * Speed ramps * Match cuts * Natural jump cuts * Handheld camera movement * Slight rolling shutter * Smartphone HDR exposure * Real travel vlog energy * No slow motion * No cinematic movie grading * Authentic TikTok creator aesthetic * Warm skin tones * High realism * Consistent face identity across every scene * 16:9 landscape

開啟結果

當所選選項支援時的音訊

某些 Veo 配置包含音訊選項。僅在所選模型和設定提供時使用它。如適用,描述所需的聲音,然後一起檢視實際的音軌和視覺效果。

Veo 適用於你的創意工作流程

相關靈感 · Seedance 2.0An intense, cinematic, slow-motion action sequence on a damp, overcast city
檢視提示

An intense, cinematic, slow-motion action sequence on a damp, overcast city street during dusk. A young Asian man in a black hoodie is in a tight, fast-paced hand-to-hand martial arts fight with a glowing, high-tech holographic or energy-based blue dynamic cyborg. The camera dynamic moves quickly, capturing sharp dodges, blocks, and misses. VFX & Transformation Sequence: The man jumps back into a defensive stance. A close-up reveals his forearm glowing with intricate, incandescent orange and gold circuit patterns, emitting light steam. With a dramatic fist clench, an explosion of golden sparks and metallic nano-particles erupts, rapidly morphing and crawling up his arm to form a high-tech gold and black metallic gauntlet. The camera pans out as the nano-armor swiftly encases his entire body, forming a sleek, studio-quality gold and black exo-suit with a fabric hood. The hero extends his arm, generating a swirling vortex of golden energy that fires a powerful concentrated beam across the street, illuminating the wet asphalt. Lighting & Atmosphere: Moody, desaturated overcast daytime lighting with a soft sunset glow in the background. The scene features highly reflective surfaces—wet ground, glossy cars (Hyundai SUV), and polished metallic armor. The hero’s helmet features a smooth, reflective gold visor that suddenly activates, glowing with sharp, vibrant orange LED eyes. Ultra-detailed, photorealistic, 8k resolution, cinematic composition, Hollywood-style visual effects.

開啟結果

產品和活動動態概念

探索產品場景在攝影機運動或簡單動作下可能呈現的樣子。當你已有起始構圖時使用圖像。在為你的活動選擇結果前,檢查產品幾何形狀、標籤和跨幀的變化。

相關靈感 · Seedance 2.0A cozy Studio Ghibli-inspired animated cooking scene of handmade dumplings
檢視提示

A cozy Studio Ghibli-inspired animated cooking scene of handmade dumplings being prepared, folded, arranged in a bamboo steamer, steamed over a bubbling pot, and served with soy sauce. Warm wooden kitchen, soft natural lighting, gentle steam, detailed food close-ups, and a fluffy orange tabby cat calmly watching from a chair before relaxing sleepily in the background. Wholesome atmosphere, cinematic camera movements, highly detailed, comforting, magical realism, 4K.

開啟結果

社群影片想法

撰寫一個只有一個主體和一個主要動作的短場景。為預期放置位置選擇支援的長寬比。聚焦的鏡頭比包含許多不相關事件的提示詞更容易評估和精煉。

相關靈感 · Seedance 2.0Ultra-realistic 4K first-person travel vlog in New York City during golden
檢視提示

Ultra-realistic 4K first-person travel vlog in New York City during golden hour. A young woman films everything herself using her smartphone in selfie mode while exploring the city with her husband. The video begins with a cheerful selfie in Times Square, then she flips the camera to capture the bustling streets, iconic yellow taxis, towering skyscrapers, and vibrant city life. She records her husband smiling, walking beside her, holding her hand, and laughing naturally. They grab coffee from a street café, cross a famous crosswalk together, share candid moments in Central Park, and admire the Manhattan skyline. The vlog ends with a happy selfie of the couple with the New York skyline glowing behind them at sunset. Authentic handheld phone movements, natural front and rear camera transitions, slight camera shake, realistic smartphone lens perspective, documentary-style, photorealistic, cinematic lighting, genuine expressions, highly detailed, no AI artifacts, no music—only natural city ambience.

開啟結果

分鏡和鏡頭探索

在建立完整序列之前測試計劃鏡頭的方向。指定構圖、運動和氛圍。將結果與你的分鏡進行比較;生成的片段是探索,而非確切連續性的保證。

相關靈感 · Seedance 2.0Ultra-cute pastel cloud animation, soft rounded chibi style, bright sky
檢視提示

Ultra-cute pastel cloud animation, soft rounded chibi style, bright sky tones, smooth bouncy physics, cinematic motion. 0–4s: Cloud floats → wakes with cute expression. 4–8s: Spins and zooms playfully, leaving puff trails. 8–12s: Interacts with sun/birds, energetic movement. 12–15s: Hero center frame, happy bounce, sparkle. Soft daylight, smooth camera, wholesome vibe, no flicker, high quality.

開啟結果

從靜態圖像產生動態

從圖片開始並描述一個克制的運動,例如緩慢推向主體或場景中的微風。隨著片段進展,檢視主體和背景的變化。

如何使用 Veo AI 影片生成器

  1. 相關靈感 · Seedance 2.0Scene & Mood: A quiet uncanny rural road scene. The open countryside from
    檢視提示

    Scene & Mood: A quiet uncanny rural road scene. The open countryside from image2 feels real and natural, but the surveillance mood of image1 leaks into the road through shallow puddle reflections. Lonely, restrained, realistic, unsettling. Frame Map: Start high and wide above the road from image2 , road running from lower foreground to deep center horizon. Green fields stretch on both sides, pylons sit far in the distance, heavy clouds fill the sky. The woman is small, lower center right, walking away from camera. As the camera lowers behind her, keep the road as the main leading line. End with her centered in the middle distance and one foreground puddle reflecting a single painted eye from image1 . Subject Lock image2 : Preserve the lone rear view woman from image2 : long dark hair, sleeveless dark patterned dress, white shoes, slim body shape, quiet walking posture, small silhouette against the landscape. Her face is never revealed. image2 keeps the same visible body shape, hair, wardrobe, scale, and silhouette throughout. Do not add bags, accessories, new clothing, or a full turn toward camera. Cross Frame Rules: Keep the woman on the same road, walking away, with stable scale and screen position. Maintain the same rural road, green fields, pylons, cloud cover, and natural daylight palette. The eyes from image1 appear only as reflections in shallow puddles, never as literal tanks or objects in the field. No extra people, vehicles, animals, fantasy glow, or signage text. Movement: Shot 1 (0–5s): High wide elevated tracking shot glides forward above the rural road. The woman walks slowly away. Fields ripple in light wind, clouds drift, pylons remain distant and still. Shot 2 (5–10s): Camera lowers behind her into a grounded rear-follow angle. Her dress hem moves softly, hair shifts in the wind, shoulders rise and fall with breath. Small puddles on the asphalt begin catching dark curved reflections like the painted eyes from image1. Shot 3 (10–15s): Camera continues a slow push. She stops in the middle of the road without turning around. Wind moves the fields in broad waves. One foreground puddle sharpens into a clear reflected painted eye, aligned beneath her path. Camera settles into a locked final hold. Last Frame: At 15 seconds, the woman stands still in the middle distance, back to camera, centered on the road under heavy clouds. Green fields stay wide and quiet on both sides, pylons faint on the horizon. In the lower foreground, one shallow puddle holds a crisp reflected eye from image1 . Camera locked. No on screen text, no captions, no signage typography, no rendered text in the frame. World Plate: Use image2 as the main world: rural asphalt road, green fields, scattered trees, distant pylons, heavy clouds, open horizon, natural field texture, muted daylight. Use image1 only for wet road reflection mood, cool gray undertone, and weathered painted-eye detail. Road is damp, not mirror glossy. Mild haze and soft air depth. Sound Bed: Diegetic audio only: rural wind through grass, faint power line hum, soft footsteps on damp asphalt, fabric movement, distant insects, small puddle sounds, open-air room tone. No music, no lyrics, no soundtrack, no score. Capture Realism: Natural daylight, soft highlight rolloff, lifted shadows, realistic field texture, matte damp asphalt, believable puddle distortion, mild grain, no plastic shine, no AI gloss, no over sharp edges, no crushed blacks, no clipped whites. The reflected eye looks painted, weathered, and warped by water. Camera Capture: 15 seconds total runtime. 24fps, 180 degree shutter, wide latitude cinema capture, slow elevated tracking descent into grounded rear-follow movement, subtle handheld breath after 10 seconds, 35mm lens feel with slight 50mm compression in the final hold, fine 35mm grain, muted natural color grade with cool-gray cloud shadows and deep green fields.

    開啟結果

    選擇起始模式

    選擇「文字轉影片」以用文字描述場景。選擇「圖像轉影片」並加入支援的參考資料以進行圖像引導的鏡頭。從此模型系列中選擇可用的 Veo 版本。

  2. 相關靈感 · Seedance 2.0Open on a soft, romantic setting with warm ivory silk fabric and delicate
    檢視提示

    Cinematic 10-second luxury unboxing video for the "ETERNAL VOW" wedding ring by LUMINA Bridal. Open on a soft, romantic setting with warm ivory silk fabric and delicate rose-gold accents under gentle golden sunrise lighting. A sleek matte cream and rose-gold jewelry box sits centered. The lid opens smoothly with a soft click, revealing a timeless platinum and diamond wedding band nestled in blush velvet. A gloved hand in ivory silk gently lifts the ring. The diamonds sparkle with warm brilliant fire as soft golden particles and elegant light ribbons swirl around it. The ring levitates gracefully, slowly rotating to showcase its perfect round brilliant cut and engraved inner band, while delicate golden lens flares and sparkling highlights dance across the frame. Smooth cinematic camera movement: gentle orbit around the floating ring with a slow upward tilt. Photorealistic details, ultra-premium commercial quality, volumetric lighting, shallow depth of field, luxurious warm rose-gold and ivory color grading, hyper-realistic textures, romantic and timeless mood.

    開啟結果

    撰寫鏡頭並檢視其報價

    描述主體、動作、場景、光線和攝影機運動。選擇你的模型提供的時長、構圖、解析度和音訊選項。在生成前檢視顯示的點數費用。

  3. 相關靈感 · Seedance 2.0Ultra-detailed martial arts character design sheet, Master Chen , legendary
    檢視提示

    Ultra-detailed martial arts character design sheet, “Master Chen”, legendary Wing Chun grandmaster, cinematic concept art presentation, premium AAA game character sheet, realistic stylized Asian kung fu master, traditional Chinese martial arts aesthetic, minimalist ink-wash inspired layout, clean white textured background.Character description: An elderly Chinese Wing Chun master around 72 years old, calm and disciplined expression, long white beard, white eyebrows, weathered wise face, hair tied in a traditional topknot bun, lean martial artist physique, strong posture, piercing eyes full of wisdom and precision. Wearing elegant black traditional kung fu uniform (Tang suit / changshan inspired), loose martial arts pants, cloth kung fu shoes. Hands posed in classic Wing Chun guard stance. Style:realistic stylized concept art wuxia-inspired visual design high-end video game concept sheet cinematic martial arts aesthetic Chinese ink painting influence subtle watercolor texture grounded realistic anatomy elegant minimal composition monochrome earthy palette with black, gray, beige, brown tones ultra detailed facial textures realistic cloth folds dramatic soft lighting premium artstation quality ghost of tsushima + sifu + ip man inspired visual mood Character sheet includes:full body hero pose front / side / back turnaround views facial expression detail panels costume close-ups hand detail studies shoe detail Wing Chun wooden dummy training equipment martial arts technique silhouettes philosophy/profile section color palette swatches Chinese calligraphy elements professional layout typography concept art presentation board Profile details: Age: 72 Height: 172 cm Discipline: Wing Chun Role: Master / Instructor Personality: Calm, wise, disciplined Focus: Efficiency, timing, control Philosophy: “Less is more. Flow is power.”Mood keywords: wisdom, discipline, mastery, balance, simplicity, precision, inner peace, martial philosophy, traditional kung fu, eastern minimalismNegative prompt: anime, cartoon, sci-fi armor, cyberpunk, low quality, blurry, bad anatomy, extra fingers, exaggerated muscles, modern casual clothing, colorful neon palette, messy layout, flat lighting, childish style, unrealistic proportions, low detail, overexposed, noisy texture, fantasy monster elements

    開啟結果

    生成並檢視片段

    無需註冊即可提交要求,並在歷史記錄中檢視其狀態。觀看完成的短片,留意動態、主體變化及適用時的音訊。下載結果或為您的下一個鏡頭修改提示詞。

撰寫更清晰的 Veo 提示詞

相關靈感 · Seedance 2.0Create a premium, modern high-fashion CHARACTER CONCEPT ART SHEET in a 16:9
檢視提示

Create a premium, modern high-fashion CHARACTER CONCEPT ART SHEET in a 16:9 widescreen layout on a pure white background, following this exact structured composition and visual aesthetic: [STYLE]: high-contrast digital anime illustration with clean, slightly textured graphite outlines, sharp cel-shaded shadows, and soft hand-painted watercolor textures. The style should feel like premium, highly fashionable dark-fantasy concept art, featuring a bold, expressive color-clash harmony of deep charcoal gray, vibrant toxic-pink, tattered crimson, and tarnished gold. [SUBJECT_DESCRIPTION]: A powerful and terrifyingly corrupted, monstrous female demi-god. She wears an elegant but tattered, flowing crimson-and-charcoal robe that drapes gracefully over her form, revealing decaying lace details and worn leather bindings. Her left arm and legs are organic, skeletal structures made of tarnished gold and bronze gothic armor plating fused with twisting wooden roots and brambles. Sprouting from her back are huge, semi-decayed spectral moth-like wings made of shimmering toxic-pink and gold pollen. Her face is partially obscured by a golden porcelain crown-like mask, revealing long, wild crimson hair. She stands with a majestic, menacing posture, exuding a beautiful yet decaying, god-like authority. Layout Composition (Organize strictly into these sections): 1. LEFT PANEL: METADATA & TURNAROUND - Large, bold vertical typography of the character's Name in a clean, heavy sans-serif condensed font. - Metadata Block: "ROLE", "CORE MOOD", and "VISUAL SIGNATURE" detailed in a clean sans-serif font. - MINI TURNAROUND: 3 small, clean vertical figures showing "neutral", "back view", and "profile" stances. - SILHOUETTE STUDY: 3 small, solid-black silhouette matchings of the turnaround stances underneath. - EXPRESSION STUDY (Bottom Left): Exactly 4 small, close-up headshots showing subtle shifts in expression (e.g., Neutral, Pensive, Confident, Smile). 2. CENTRAL PANEL: THE CENTERPIECE - One giant, dominant full-body figure of the character standing in a cool, relaxed signature pose (e.g., hands in pockets, looking down/aside). This figure must visually dominate the center of the sheet with maximum detail. 3. RIGHT PANEL: MULTI-POSE & THEMATIC STUDIES - POSE STUDY: Exactly 4 to 5 small-to-medium figures in dynamic, casual stances (such as "seated", "crouching", "leaning", "low angle", or "top view"), each with a small handwritten-style descriptive label underneath. - OPTIONAL SPECIALTY STUDY: A vertical strip of 4-6 artistic, painterly portraits with expressive brushstroke or paint-splatter borders showing morphing, stylized, or glitch-like variations of the character's face. 4. BOTTOM RIGHT PANEL: DETAIL STUDY - Exactly 5 small square close-up crops highlighting the character's unique features, accessories, and textures (e.g., fabric weave, jewelry, specific utility pockets, unique footwear soles). - Each crop must have a clean, neat handwriting-style label underneath describing the detail (e.g., braids medium length, layered fit, chunky sole). Keep the entire presentation visually consistent, clean, and highly professional, mimicking a real studio-grade character design bible.

開啟結果

主體、動作和場景

給鏡頭一個中心主體和一個清晰的動作。例如:「一位自行車手在日出時經過一間安靜的咖啡館。暖光反射在窗戶上。一個平滑的追蹤鏡頭跟隨騎手。」將不同場景放在不同請求中。

相關靈感 · Seedance 2.0A wide cinematic landscape shot of a large outdoor concrete staircase beside
檢視提示

A wide cinematic landscape shot of a large outdoor concrete staircase beside a massive modern building covered with tall rectangular glass panels. Warm late afternoon sunlight creates realistic golden highlights and long shadows across the stairs and open concrete plaza. A broad sloping silver metal handrail runs diagonally down the center of the staircase toward the camera. A young female roller skater stands near the upper-right side of the stairs, wearing an oversized mustard-yellow sweatshirt with a red graphic on the chest, fitted black pants, a white baseball cap, and white quad roller skates with subtle pink details. Her long brown ponytail moves naturally in the breeze. 0–3 seconds: One continuous stationary smartphone-style wide shot. The skater rolls carefully into position at the top of the sloping rail, looks down its length, bends her knees and raises her arms, preparing for the stunt. A few distant people sit casually on the stairs in the background. 3–6 seconds: She gathers confidence, rolls forward and crouches lower. Her wheels vibrate realistically over the stone surface. She jumps upward, turning her body slightly sideways, attempting to land both roller skates on top of the wide rail. 6–9 seconds: She lands in a deep crouch on the rail and begins sliding downward. For a brief moment the stunt appears successful. Her arms stretch outward for balance, her sweatshirt flutters, her ponytail swings and the roller skates scrape realistically against the metal. 9–12 seconds: Her weight suddenly shifts to one side. One skate slips from the rail, causing her hips to rotate awkwardly. She grabs at the rail but cannot recover. She slides sideways, tumbles off the lower end and falls onto the flat concrete plaza. Her white cap flies off and bounces nearby. Use convincing gravity, momentum, body weight and natural protective arm movements. 12–15 seconds: She completes a small shoulder roll and ends lying on her side beside the staircase, one knee bent and one leg extended. Her loose hair spreads across the ground while the cap rests a short distance away. She remains still for a moment, then gives a tiny embarrassed movement showing that she is unharmed. The camera continues recording without cutting. Photorealistic human anatomy, authentic roller-skating movement, physically accurate fall, realistic fabric and hair simulation, natural facial reactions, documentary style exposure, subtle handheld micro shake, crisp environmental detail, believable sunlight and shadows, real-world motion blur, 30 fps, no cinematic slow motion, no camera cuts. Audio: quiet outdoor city ambience, roller wheels rumbling over stone, brief metal scraping sound during the slide, clothing movement, a realistic soft impact and the cap tapping against the pavement. Avoid: serious injury, blood, exaggerated violence, impossible acrobatics, floating body, rubber limbs, distorted hands, duplicated skates, changing clothes, morphing rail, disappearing cap, artificial CGI appearance, dramatic camera movement, zooming or scene transitions.

開啟結果

攝影機運動和構圖

指定廣角鏡頭、特寫或其他構圖選擇,然後描述一個簡單的運動。避免矛盾的攝影機方向。對於圖像轉影片,解釋鏡頭應如何從參考資料發展。

相關靈感 · Seedance 2.0D cinematic animation, stylish museum heist comedy-action, glossy lighting,
檢視提示

Style: 3D cinematic animation, stylish museum heist comedy-action, glossy lighting, playful suspense, expressive characters, 15-second vertical video pacing, 7 quick cuts [00:00-00:02] Wide 3D shot inside a grand museum gallery at night, with marble floors, tall columns, velvet ropes, glass display cases, and a valuable painting glowing under a spotlight on the far wall. A sneaky thief in a black outfit quietly enters from the shadows, tiptoeing with exaggerated confidence. SFX: quiet museum ambience, soft footsteps, suspenseful pizzicato strings. [00:02-00:04] Low-angle shot as red security laser lines activate across the gallery like a glowing web. The thief freezes, then smirks, raising one eyebrow as if the challenge is easy. Reflections of the lasers shimmer across the polished floor. SFX: laser hum, tiny electronic beeps, playful suspense music. [00:04-00:06] Dynamic side tracking shot as the thief moves through the laser grid with feather-light motion, bending backward, sliding sideways, stepping on one toe, and twisting past the beams with dancer-like precision. His coat barely misses the lasers. VFX: glowing red laser trails, slow-motion dust particles, soft floor reflections. SFX: fabric swish, soft whooshes, light comedic drum taps. [00:06-00:08] Close-up of the thief’s face as he finally reaches the painting, looking proud and whispering, "Too easy." He gently places one hand on the frame. The painting instantly flashes with a hidden sensor light. SFX: tiny click, sudden electronic chirp, suspense music cuts. [00:08-00:10] Fast wide shot as the entire museum alarm system erupts. Red lights flash, sirens scream, and metal security shutters begin lowering. The thief’s confident smile collapses into panic. He looks around and says, "Okay... not easy." VFX: flashing red alarm lights, blinking sensors, light haze. SFX: loud alarm, warning beeps, metal shutters. [00:10-00:12] Fast handheld 3D chase shot as guards rush into the gallery from both sides with flashlights, pointing toward the thief. He tries to step back, but the laser grid reactivates behind him, blocking his escape. He slowly raises his hands with a nervous smile. SFX: running footsteps, flashlight clicks, guard shouts, alarm continuing. [00:12-00:15] Final comedic wide shot as two guards calmly surround the thief while the painting remains safely on the wall. The thief stands trapped under the spotlight, hands raised, still holding only a tiny loose piece of tape from the frame. One guard says, "Nice dancing." The thief sighs and replies, "Thank you." End on the thief looking embarrassed as the red alarm lights flash around him. SFX: siren fading slightly, comedic bass hit, final playful music sting.

開啟結果

檢視運動和聲音

從頭到尾檢查片段,而不是只判斷其第一幀。當音訊受支援並被選中時,描述預期的聲音並檢視實際生成的內容。圍繞你觀察到的問題精煉下一個提示詞。

Veo 問題解答

我可以將 Veo AI 影片生成器與文字和圖像一起使用嗎?

可以,此頁面為已啟用的 Veo 系列整合了文字轉影片和圖像轉影片。所選模型決定哪些報價模式和設定可用。圖像轉影片需要支援的參考圖像。

圖像轉影片是指上傳影片嗎?

不是。圖像轉影片從靜態圖像開始並生成片段。此頁面不提供影片轉影片編輯。使用圖像選擇器作為參考,並描述你想要的運動。

我應該如何描述攝影機運動?

說出構圖和一個主要的攝影機運動,例如緩慢推近或追蹤鏡頭。分別描述主體動作。保持運動具體有助於你將片段與預期鏡頭進行比較。

Veo 在此支援什麼時長和解析度?

可用選項會為你選擇的 Veo 版本顯示。它們取決於活動的模型配置和報價組合。此頁面不承諾生成器中未出現的時長或解析度。

Veo 可以為我的影片生成音訊嗎?

音訊取決於所選的 Veo 版本及其可用設定。如果提供音訊選項,選擇它並在相關時在提示詞中包含所需的聲音。檢視生成的音軌;並非每種配置都保證有音訊。

點數如何計算?

顯示的報價使用所選模型、模式、時長、解析度和其他支援的選項。在提交前檢查金額。前往定價頁面比較可用方案和點數購買。

完成的影片會出現在哪裡?

登入後,在歷史記錄或資產中檢視完成的片段。開啟結果以播放它並使用可用的下載控制項。你可以修改描述或參考資料並提交新請求以獲得另一個變化。

用 Veo 製作你的下一部影片

將你的構想帶到 CanvasSeed,選擇你的起始模式,並在生成前檢視可用設定。

使用 Veo 創作