LogoAhaPrompt
  • 提示詞
  • 精選
  • 模型
  • AI 反推
  • 定價
  • 投稿
  • 部落格
AhaPrompt

圖片、影片、音訊 AI 提示詞庫

探索

  • 全部提示詞
  • 精選提示詞
  • 圖片提示詞
  • 影片提示詞
  • 音訊提示詞
  • AI 模型

工具

  • AI 反推工作臺
  • 圖片反推提示詞
  • 影片反推提示詞

資源

  • 定價方案
  • 投稿提示詞
  • 部落格

版權所有 © 2026 AhaPrompt 保留所有權利。

隱私政策條款與條件退款政策版權申訴 / DMCA
首頁AI 提示詞庫AI 影片提示詞角色替换团块

角色替换团块

KuaishouKling 4.0Kuaishouvideo

视频1:将色块替换为图像1中的生物。匹配色块的大小。其他一切保持不变。

詳細資訊
类型
AI 影片提示詞
畫面比例
16:9
Duration
0:57
更新於
2026年11月1日
分類
VFXAnimationFantasy
Source

如何使用這條提示詞

  1. 1點選上方複製按鈕,複製完整提示詞。
  2. 2開啟 Kuaishou 的 Kling 4.0,或你常用的 AI 影片生成工具。
  3. 3貼上提示詞,保留鏡頭與運動描述詞——它們決定了畫面的運鏡效果。
  4. 4按示例參數(16:9 · 0:57 · VFX · Animation · Fantasy)生成,再小步迭代——措辭的細微變化會顯著影響結果。

常見問題

這段影片是用哪個 AI 模型生成的?

本片段由 Kuaishou 的 Kling 4.0 生成,頁面上的提示詞就是生成它的原始文字。

這條提示詞用了哪些參數?

示例由 Kling 4.0 生成,參數:16:9 · 0:57 · VFX · Animation · Fantasy。沿用這些參數可獲得最接近的效果。

怎樣才能生成相似的效果?

先原樣複製提示詞,只改主體、保留鏡頭/運動/光影描述。影片模型對措辭很敏感,多試幾次為佳。

可以修改這條提示詞嗎?

當然可以。提示詞最適合作為起點——替換場景、調整節奏詞或寬高比,適配你的專案。

更多 Kling 4.0 提示詞

0:15
KuaishouKling 4.0
FORMAT Create a premium Pixar-style 3D animated bathroom comedy using the supplied video as the direct performance reference. Match Video 1's full duration, approximately 14.5 seconds, vertical 9:16 composition, playback speed and original action sequence. Replace only the two characters and make the minor bathroom decor changes specified below. REFERENCE ROLES Video 1: master reference for the complete performance, action timing, facial expressions, body movements, prop handling, spatial composition, camera behavior and audio. Image 1: appearance reference for the replacement woman. Image 2: appearance reference for the replacement man. Transfer the original woman's performance to the woman from Image 1. Transfer the original man's performance to the man from Image 2. The photos control character appearance and clothing only. Do not copy their neutral standing poses, studio backgrounds or full-body framing into the video. PRIORITY Use Image 1 and Image 2 for character identity, hairstyle, colors and clothing. Use Video 1 for all actions, expressions, timing, framing and interaction. Only the bathroom finishes may change as described below. Do not reinterpret the scene or invent additional actions. WOMAN The same beautiful 25-year-old adult woman from Image 1. Preserve her exact face, green eyes, freckles, skin tone and adult proportions. Preserve her pastel-lavender high wavy ponytail, including its volume, length and loose face-framing strands. Preserve her ivory long-sleeve henley, sage-green lounge trousers and cream slippers wherever visible. Do not replace her ponytail with the original woman's dark bun. Her facial expression must follow the video performance, not remain fixed in the reference photo's smile. MAN The same 25-year-old adult man from Image 2. Preserve his exact face, hazel eyes, tousled dark-brown wavy hair, eyebrows, light stubble and adult proportions. Preserve his navy short-sleeve T-shirt, taupe lounge trousers and dark-gray slippers wherever visible. His expression must follow the video performance, not remain fixed in the reference photo's smile. EXACT PERFORMANCE MATCH Use Video 1 as the continuous motion and expression guide from its first frame to its last. Match each character's: - Head position, tilt, turn and forward lean. - Gaze direction, blinks, eyelid movement and eye widening. - Eyebrow movement, mouth opening, lip pursing, smiles and annoyed reactions. - Shoulder movement, arm trajectory, wrist rotation and finger gesture. - Hand-to-face contact and interaction with grooming tools. - Start and end timing of every gesture. - Original brief pauses and reaction delays. Preserve the original timing difference between the woman's actions and the man's imitations. Do not make them move in perfect synchronization unless they do so in the source. The man imitates only at the moments shown in Video 1. At other moments, reproduce his watching behavior and facial reactions. Do not invent continuous mirroring. Preserve the full movement paths, not merely a few key poses. Keep the action flowing naturally between them. Do not add pauses, accelerate gestures, prolong reactions or repeat completed movements. ACTION ANCHORS These anchors identify the source actions, not a replacement timeline. Their timing, order, duration and detailed execution must come directly from Video 1. Opening: The woman applies mascara near her eye with the original wide-eyed, open-mouth expression. The man stands behind her toward screen-right and imitates her using his empty hand. Reproduce the exact moment when she lowers the wand and reacts. Middle: Match her downward glance and displeased expression, his amused reaction and the original transition from mascara to the blue razor. Reproduce her grooming movements around the chin, cheek and upper-lip area. Match how her free hand stretches her cheek and how her mouth changes shape. Preserve the man's original watching expressions throughout. Ending: Reproduce his finger gesture near his own upper lip, her reaction, their final gaze shifts and her reach toward screen-right. End on the same phase of her movement toward the counter as the source. Do not add a new ending, punchline, embrace or final pose. CAMERA AND SPATIAL COMPOSITION Match Video 1's camera position, perspective, framing and near-static composition. Keep the woman in the foreground and the man behind her, offset toward screen-right, with the same relative screen positions and apparent sizes as the source. Preserve visibility of their faces and hands at the same moments as in Video 1. Do not widen to a full-body view because the character photos show full bodies. Preserve the original screen orientation and hand assignments. Do not horizontally flip the scene. Reproduce only the mirror presentation already visible in Video 1. Do not add a new mirror, a reverse angle, an over-the-shoulder view or a second reflected copy of either character. No visible camera or camera operator. Follow the source camera behavior and any source edits. Do not introduce additional cuts, zooms, pans or camera shake. BATHROOM Keep the original room layout and functional geometry: - Vanity counter in the foreground. - Wooden storage toward screen-left. - Window toward screen-right. - The same available space for grooming gestures. - Essential props in positions consistent with the original hand movements. Make only these subtle decorative changes: Muted sage-green wall tiles, pale oak cabinetry and a warm ivory stone countertop. Keep the original arrangement of grooming bottles, dispensers and the toothbrush cup sufficiently similar that every reach and prop interaction remains logical. Do not add objects that obscure faces or interfere with ha
檢視詳情
0:30
KuaishouKling 4.0
The Magical Hidden World – 30 Seconds Create a breathtaking, ultra-realistic cinematic fantasy film, exactly 30 seconds long, in vertical 9:16 format. Scene 1 (0–5 sec): A young woman walks alone through a mysterious ancient forest at sunrise. Golden sunlight passes through giant trees, soft fog covers the ground, cinematic wide shot. Scene 2 (5–10 sec): She discovers a glowing magical doorway hidden behind a giant waterfall. Tiny glowing particles float through the air. Scene 3 (10–15 sec): She slowly walks through the magical doorway and discovers a breathtaking floating world with waterfalls, crystal mountains, and glowing flowers. Scene 4 (15–22 sec): A giant beautiful white dragon flies gracefully across the sky above floating islands. The camera follows its movement with a sweeping cinematic aerial shot. Scene 5 (22–30 sec): The woman stands on a cliff overlooking the magical kingdom as the sky transforms into beautiful golden and purple colors. The camera slowly pulls back, revealing the enormous fantasy world. Visual Style: Photorealistic, cinematic fantasy, 4K, Unreal Engine quality, realistic environments, atmospheric lighting, smooth camera movements, stunning details, natural physics. Negative Prompt: No products, no logos, no text, no watermark, no distorted faces, no extra limbs, no flickering, no cartoon appearance.
檢視詳情
0:30
KuaishouKling 4.0
Create a 30-second ultra-realistic cinematic video of a breathtaking underwater kingdom. A beautiful woman wearing an elegant shimmering blue gown walks through a crystal palace beneath the ocean. Giant whales glide past glowing glass towers, colorful fish swim around her, and magical jellyfish illuminate the water. She touches a glowing pearl, awakening an enormous ancient sea dragon that rises majestically behind the palace. End with a spectacular wide-angle shot of the underwater kingdom glowing beneath the sunlit ocean. Photorealistic details, cinematic camera movement, realistic water physics, stunning VFX, smooth motion, consistent character appearance, dramatic lighting, professional Hollywood quality, vertical 9:16, 30 seconds.
檢視詳情
0:15
KuaishouKling 4.0
This is a funny skit video going viral. Replace the middle character in Video 1 with the man in Image 1. Keep everything else the same, including the background, camera angles, movements, expressions, timing, and actions. Ensure the replacement looks natural and blends seamlessly into the original video.
檢視詳情
0:30
KuaishouKling 4.0
CHARACTER LOCK: @image1 = Lala. Preserve exact face, identity, hairstyle, body, outfit and accessories. Her magical doll remains present and subtly responds to her gestures. @image2 = Cerberus. Preserve its exact three-headed anatomy, horns, proportions and appearance. Each head moves independently. ENVIRONMENT & SCALE: An enormous ancient volcanic caldera with black obsidian rocks, magma fissures, ash, steam and distant mountains. Lala stands on an elevated plateau; Cerberus occupies a lower basin 100 meters away. Lala is 1.7m tall and Cerberus 5.1m. Maintain their scale, distance and positions. Lala never leaves the plateau. 00–04 | CONFRONTATION: Aerial camera reveals the caldera, Lala on the plateau and aggressive Cerberus in the lower basin. Its three heads roar independently as its paws crush volcanic rocks. Lala remains calm. 04–07 | DOUBLECAST: Camera orbits Lala as she activates Doublecast through controlled hand gestures. Her doll responds subtly. No spoken announcement. 07–13 | BLIZZAGA: Lala performs elegant, flowing gestures and shouts 「ブリザガ!」 (Burizaga). A powerful frost wave travels across the battlefield, freezing the basin, rocks and Cerberus. Ice progressively covers its entire body and all three heads, immobilizing it completely. Camera tracks the advancing frost and reveals the frozen beast. 13–18 | FIRAGA: Lala immediately changes to firm, commanding gestures and shouts 「ファイガ!」 (Faiga). Magma erupts through ground fractures, lava flows across the basin, and existing ice melts into thick steam. The massive ice prison around Cerberus remains intact. Do not free or thaw it. 18–22 | ELEMENTAL CONVERGENCE: A violent vortex forms around Cerberus. Sharp ice shards and heavy flowing magma converge from multiple directions, violently shattering its ice prison and striking the creature. Keep Cerberus recognizable at the center. No additional spell announcement. 22–30 | ULTIMA: Lala performs a distinct authoritative gesture and shouts 「アルテマ!」 (Arutema). A colossal black-violet gravitational sphere forms above the basin, distorting space and pulling in ash, ice and debris. It descends and detonates directly over Cerberus. A devastating radial shockwave destroys the basin, displacing magma, ice and massive debris. The explosion remains confined to the lower basin, leaving Lala's elevated plateau outside the main blast. The sphere dissolves into fading particles. End on a wide shot of Lala calmly overlooking the destruction. CAMERA & STYLE: Premium live-action fantasy cinema, realistic elemental physics, cinematic lighting and sophisticated VFX. Use aerial wides, medium-full casting shots and dynamic ground-level tracking. Preserve clear geography and cause-and-effect. STRICT RULES: Exact identities and anatomy, consistent scale and distance, persistent environmental damage, distinct spell gestures. Cerberus stays frozen until convergence. No teleportation, duplicate characters, extra creatures, disappearing effects, resets, cartoon CGI, random cuts or unrelated dialogue.
檢視詳情
0:30
KuaishouKling 4.0
Create a beautiful 30-second Pixar-inspired 3D animated short featuring a sweet little girl wearing a pastel-yellow raincoat, pink knitted scarf and brown rain boots. She has big expressive eyes and brown hair in two ponytails. Scene 1 (0–5s): A Little Discovery On a rainy woodland path filled with glowing mushrooms and wet flowers, the girl discovers a tiny baby hedgehog shivering beneath a fallen leaf. She kneels down and gently offers her hands. Scene 2 (5–10s): A Gentle Rescue The hedgehog uncurls. She carefully picks it up, shelters it beneath her raincoat and carries it to a cozy wooden cottage. Scene 3 (10–16s): A Warm Welcome Beside a glowing fireplace, she places the hedgehog in a soft towel-lined basket, gently dries its fur and offers fresh water. Show adorable close-ups of its tiny face. Scene 4 (16–22s): A New Friendship She offers the hedgehog some suitable food. After eating, it climbs out, waddles toward her slippers and gently touches her hand with its nose. She smiles and lets it snuggle beside her. Scene 5 (22–26s): Finding Home The next morning, she carries the hedgehog outside in its basket. Near a flower garden, a larger hedgehog waits beside a small wooden shelter. The baby recognizes its mother and waddles toward her. Scene 6 (26–30s): A Beautiful Goodbye The mother and baby touch noses. The girl smiles as the baby turns back for one last look before following its mother into the garden. The girl waves goodbye as the camera pulls back through the golden morning light. Fade out. Visual Style: Premium Pixar-inspired 3D animation, expressive characters, detailed fluffy fur, natural animal movements, magical woodland atmosphere, soft rain, warm interiors, beautiful lighting, smooth animation and cinematic camera movements. Audio: Rainfall, tiny footsteps, hedgehog snuffling, fireplace sounds and morning birds. No narration, dialogue, music, subtitles or text. Character Consistency: Maintain the girl's exact face, hairstyle and outfit throughout. Keep both hedgehogs visually consistent, with the mother noticeably larger. Negative Prompt: No photorealism, anime, 2D animation, scary expressions, aggressive animals, distorted anatomy, extra limbs, character inconsistency, outfit changes, unnatural movements, flickering, jitter, text, subtitles or watermark.
檢視詳情
0:30
KuaishouKling 4.0
Create an ultra-realistic cinematic adventure video, exactly 30 seconds long, in vertical 9:16 format. Scene 1 (0–6 seconds): A beautiful young woman wearing an elegant flowing white dress walks through a mysterious tropical jungle. Sunlight shines through giant trees while soft mist floats in the air. Scene 2 (6–12 seconds): She discovers an enormous waterfall hidden behind ancient mountains. Thousands of birds fly across the sky as water crashes dramatically into a crystal-clear lake. Scene 3 (12–18 seconds): She approaches the edge of the lake and touches the water. Magical glowing flowers begin blooming around her feet. Scene 4 (18–24 seconds): The camera rises above the jungle, revealing breathtaking floating islands, multiple waterfalls, colorful birds, and magnificent mountains covered in clouds. Scene 5 (24–30 seconds): A giant rainbow appears above the waterfall. The woman stands peacefully looking at the magical landscape as the camera slowly pulls back into an epic aerial wide shot. Visual Quality: Ultra-photorealistic, cinematic filmmaking, 4K HDR, realistic water simulation, natural human movements, beautiful lighting, detailed environments, smooth camera transitions, professional color grading. Negative Prompt: No cartoon style, no distorted faces, no extra fingers, no unnatural movements, no flickering, no text, no subtitles, no logos, no watermark
檢視詳情
0:30
KuaishouKling 4.0
Photorealistic Hollywood live-action dark fantasy. @image1: preserve exact male identity, face, black hair, supernatural iris, facial markings, black costume, gloves, boots and fabric. SETTING: Stormy late-night Shinjuku, Tokyo. Real skyscrapers, roads, traffic, pedestrians and rooftop machinery. Natural anatomy, skin and cinematic lighting. No anime, cyberpunk or game look. All supernatural energy is blood-crimson, never purple. Seven rotating 3D geometric rings, gravitational lensing, volumetric light and invisible shockwaves. The living city reacts physically: windows crack, vehicles move, debris lifts, crowds flee and buildings fracture. 00–05: Macro rotating supernatural eye, pulsing facial markings, wind-blown profile. Crash zoom to aerial wide: hero levitates above Shinjuku as gravitational pressure shakes buildings and traffic. Spiral camera ascent. 05–11: He raises hands; crimson filaments form seven rotating rings. 180° camera orbit. Cold Japanese whisper: 「この世界は、もう終わりだ。」 Pressure wave shakes the city. 11–18: Rings expand around him. Wrist rotates; massive shockwave races through streets. Vehicles lurch, glass breaks, signs twist, debris rises, pedestrians flee. Rapid eye, finger, glass and crimson-symbol inserts. Inverted camera orbit; rings contract toward palms. 18–24: Close-up: calm expression becomes furious. Whisper: 「すべて……」 Rings compress into a singularity. Explosive shout: 「ぶっ壊してやる!!」 Hands separate; aggressive 270° orbit. 24–30: Crimson eye ignites. Primal Japanese roar: 「うおおおおおおおおッ!!!」 Arms thrust downward; rings collapse into one point. Colossal crimson energy column strikes Shinjuku. Asphalt fractures, vehicles lift, windows shatter, buildings rupture and debris cascades outward. Slow-motion impact, then explosive speed ramp. Aerial reveal of immense circular destruction zone. Final rotating iris macro; hard cut black. CAMERA: Macro inserts, smash cuts, crash zooms, spiral ascent, 180°/270° orbits, inverted tracking, aerial descent, street wides and speed ramps. Preserve sequence and spatial continuity. AUDIO: Deep Japanese male baritone, precise lip-sync, escalating rage, breathing, scream, wind, alarms, glass, metal and deep impacts. Exactly 30 seconds. Realistic physics, debris inertia, cinematic motion blur and seamless Hollywood CGI. No random cuts, flat effects or weightless destruction.
檢視詳情
瀏覽全部 Kling 4.0 提示詞探索AI 影片提示詞