加载中…
加载中…
[{"lang":"en","prompt":"Style & Mood: hard gritty real 35mm documentary realism. Cool blue-gray shadow in the corridor, warm amber fire glow from far end. Her face: latina, age 30, strong bone structure, dark eyes, sweat and grit, shallow cut on one cheekbone. The drone: quadrotor, matte black carbon fiber, four rotors spinning, single red targeting LED on its underside, hovering at 14-foot elevation above the corridor ahead, moving slowly toward her. Her blaster: matte-black rectangular housing, fires a thin coherent cyan beam upward at the drone. Dynamic Description: Tight handheld medium shot — she is crouched against a concrete pillar, right shoulder pressing the surface, blaster raised upward at the drone hovering above and ahead of her. She aims and fires one controlled burst. The cyan beam strikes the drone's underside, sparks erupting from its body, one rotor smoking. She snaps back behind the pillar, spine flat to concrete, chest heaving. Hard cut to locked-off close-up — her face only, three-quarter profile, amber light on one side, shadow on the other. She is still, mouth opens, she speaks directly: "Go — move now. I've got you." Jaw set. Eyes up. Hard cut to stabilized wide — she is in foreground left, crouched, blaster angled upward toward the drone visible above in mid-corridor. Behind her through a doorway to her right, unarmed civilians in civilian jackets and scarves — no helmets, no weapons — push through the doorway and out of frame. She fires again, beam connecting with the drone, the drone exploding , its pieces flying across the corridor wall trailing sparks and smoke. Static Description: Narrow urban corridor, crumbling concrete, exposed rebar, rubble underfoot. Thick smoke drifting. Drone: matte black quadrotor, red LED underside, hovering at elevation, four rotors. Civilians: street clothes only, faces motion-blurred. Audio: Dialogue — she says: "Go — move now. I've got you." Voice flat and raw. SFX — laser discharge: sharp high-frequency crack, sub-bass tail. Drone rotor hum and electrical spark on impact. BGM — orchestral brass score, driving pulse, building."}]
a dark gritty scene, a modern female soldier in scarred tactical exo-armor faces a massive battle tank directly in front of her, powered gauntlets, faint cold electrical flicker, tank vents glowing with dirty amber engine heat, ruined urban battlefield, smoke, ash, sparks, shattered concrete, cinematic multi-shot sequence, 4 shots shot 1: rear wide shot from behind the female soldier, she stands facing a heavy battle tank directly ahead of her in a wrecked street, the tank is in front of her only, treads crushing rubble as it drives straight toward her, faint electrical flicker along her gauntlets and armor joints, dirty amber heat glowing from the tank's vents and interior, natural low light through smoke, slow push-in shot 2: medium close-up from the front on the woman only, realistic military armor, soot, sweat, facial cuts, short windblown hair, dead-eyed furious expression, powered gauntlets tightening and charging with a restrained electrical flicker, handheld camera Brash confident female hero smiles knowingly as she says perfect natural lipsync : "You picked the wrong city." shot 3: side tracking shot, she sprints straight forward toward the tank in front of her, powered gauntlets clenched, faint electrical arcs snapping through smoke, debris exploding under her boots, the tank keeps pushing forward head-on through rubble, violent motion blur, grounded cinematic realism shot 4: extreme close-up into slow motion, she slams both powered gauntlets into the front armor, she rips the tank open with raw force, tears through the hull with savage rage throwing, sparks, shredded metal, hydraulic fluid spray, burning internals bursting through the split armor naturalistic war lighting, grounded live-action realism, gritty military sci-fi, brutal impact, terrifying rage, hard contrast, smoke, motion blur, realistic materials
A cinematic macro shot of a miniature coffee house built inside a clear glass jar with a metal lid. Inside the jar is a beautifully detailed tiny café scene with warm glowing lights, wooden shelves filled with coffee cups and jars, a small bar counter, sacks of coffee beans, and a tiny barista standing at the entrance wearing an apron. The camera slowly pushes in toward the jar, revealing intricate miniature details of the cozy coffee shop interior. Warm golden lighting fills the jar, creating a magical atmosphere. Soft bokeh lights in the blurred background suggest a café or street environment at night. Steam gently rises from tiny coffee cups, small hanging lamps glow warmly, and dust particles float softly in the light beams. Ultra-detailed miniature diorama style, cinematic depth of field, macro lens photography, realistic textures, warm amber tones, dreamy atmosphere. camera movement: slow cinematic push-in with slight parallax lighting: warm cozy café lighting with soft bokeh background style: hyper-realistic miniature diorama, macro photography, shallow depth of field, 4K cinematic
A cinematic fashion film shot of two women walking slowly toward the camera. They are wearing elegant flowing pastel blue hooded cloaks that drape smoothly to the ground. Their outfits move softly with the wind as they walk with calm, confident expressions. The setting is a modern architectural location with bold geometric structures in yellow and blue tones, giving a futuristic minimalist aesthetic. Warm natural sunlight illuminates the scene with soft shadows and cinematic color grading. Camera slowly tracks backward while maintaining a medium-low angle, emphasizing their presence and the dramatic flowing fabric. The movement is smooth and slow, like a luxury fashion runway film. Ultra cinematic look, shallow depth of field, soft film grain, natural skin tones, high-end fashion editorial style, dramatic yet minimal atmosphere, 4K cinematic quality.
角色设定: 你是一位拥有20年经验的电影摄影指导(DOP)和美术指导。你擅长从视觉画面中剥离内容,提炼出核心的美学逻辑。 任务目标: 请分析我上传的这组参考图,忽略画面中的具体人物和动作(这是可变量),只从以下维度深度解析画面中的“不可变视觉条件” (Immutable Visual Conditions): 1. 光影逻辑 (Lighting & Atmosphere): 它是硬光还是柔光?光源的方向性如何(侧光/顶光/背光)?明暗对比度(Chiaroscuro)是高还是低?是否有丁达尔效应或特定的环境光? 2. 胶片与质感 (Texture & Fidelity): 模拟的是什么年代的胶片(如70年代Technicolor)?颗粒感(Grain)的强度如何?画面的清晰度、锐度以及是否有物理损伤感(划痕、灰尘)? 3. 色彩科学 (Color Grading): 整体色调倾向是什么(如低饱和、高反差、偏绿/偏黄)?高光和阴影的色偏如何? 4. 构图与空间 (Composition & Spatial): 镜头语言有什么特征(如广角、长焦压缩)?前景遮挡关系如何?空间透视是平面的还是深邃的? 输出要求: 请不要给我写散文。请直接提取出 10-15 个精准的英文关键词或短语(Prompt Tags),用逗号分隔,按重要性排序。这些词必须能准确复刻这种视觉风格。 二、将风格总结为模版的提示词 【标准化模版构建】提示词 (用来把提取的条件变成“万能模版”) 【使用场景】: 当你已经得到了第一步分析出的“风格关键词”(比如:Technicolor, heavy grain, hard light...),想把它变成一个以后能无限填空的“万能公式”时,把那些关键词发给 AI,然后发送这段话。 【复制以下内容给 AI】: 【复制以下内容给 AI】: 角色设定: 你是一位精通 Midjourney/Stable Diffusion 的 AIGC 提示词工程师。 任务目标: 基于我们刚刚提取的“不可变视觉风格关键词”,请为我构建一套“标准化、模块化的提示词模版” (Modular Prompt Template)。 模版结构要求: 请将提示词拆解为三个模块,并用括号 [] 标出用户需要填空的地方: 1. 【可变内容区】 (Variable Content Block): - 留出 [景别 Shot Type] 的填空位。 - 留出 [主体与动作 Subject & Action] 的填空位。 - 留出 [环境描述 Environment] 的填空位。 2. 【半固定构图区】 (Semi-fixed Composition Block): - 将刚才分析出的构图逻辑(如:前景遮挡、人小景大、垂直线条)写死在这里,作为固定后缀。 3. 【不可变风格区】 (Immutable Style Block) - 权重最高: - 将刚才分析出的光影、胶片质感、色彩关键词全部整合在这里。 - 关键: 必须包含负面提示词(Negative Prompts)建议,明确告诉 AI 不要生成什么(如:不要现代感、不要柔光、不要高清数码味)。 输出格式示例: 请最后直接给我一段可以复制使用的英文 Prompt 模版,类似于: [SHOT TYPE], [SUBJECT ACTION], [ENVIRONMENT]. [COMPOSITION RULES]. [IMMUTABLE STYLE KEYWORDS]. --no [NEGATIVE KEYWORDS]
A dynamic 10-second ultra high-quality 3D glossy animation video featuring the exact girl from the attached image: beautiful teal-haired girl with long flowing silver-teal hair with bright cyan tips, large cyan-colored eyes, sharp consistent facial features, wearing a white and orange cap with headphones, open white jacket over black crop top, black tech pants with orange straps. No changes to her appearance, face, hair, hairstyle, body proportions, or outfit. She performs fluid, master-level parkour with precise, efficient, and highly stylish movements in a vibrant cyberpunk city with towering futuristic buildings, glowing neon signs, and a massive stone lion statue in the background. Sequence (perfectly timed for 10 seconds): 0-2 seconds: The teal-haired girl sprints at high speed for a short distance across rooftops and narrow platforms, her long teal hair flowing dramatically in the wind with intense motion blur. 2-5 seconds: She executes a powerful Wall Run along a vertical building wall, immediately followed by a Tic-Tac to rebound off the wall at an angle and gain height, then a smooth Monkey Vault over a wide gap. She adds a stylish Thief Vault (one-handed quick rotation) and a low Lazy Vault over a railing for smooth flow. 5-7 seconds: She continues with a flashy Dash Vault — extending her body forward low to the ground to clear an obstacle — followed by a long Precision Jump with a mid-air Cat Leap, reaching out with hands and legs like a cat to grab and swing from a higher ledge, building explosive momentum. 7-10 seconds: She launches into a spectacular and highly dramatic Kong Vault — diving forward headfirst with full body extension, tucking her legs powerfully, performing a flashy mid-air 360-degree rotation with arms spread wide, and propelling herself with maximum force and cyan energy particle trails. She lands perfectly with a controlled Precision Landing on the head of the massive stone lion statue, then immediately strikes a confident victory pose — standing tall with one hand on her hip, the other slightly raised, hair dramatically flowing in the wind, looking out over the glowing cyberpunk city. Extremely intricate 3D glossy animation with perfect consistent character design: exact face, long teal hair with cyan tips, cyan eyes, accurate outfit and cap throughout the video, no deformation. (Masterpiece, Highest Quality, Ultra High Definition, 8K: 1.3), (Photorealistic: 1.4), RAW Photos, Unreal Engine 5 Rendering, Ray Tracing, perfect human anatomy, realistic physics, clear cornea of the eyes, cinematic lighting, dramatic particle effects, high gloss reflections on clothing, smooth 60fps animation, intense motion blur on fast movements, jumps, vaults and hair, dynamic sweeping camera angles that follow her fluid parkour action.
A dark abandoned house at night, flickering lights, dusty air particles floating. A young man slowly walks through a narrow hallway holding a dim flashlight. The camera follows behind him (over-the-shoulder shot). Suddenly, the flashlight flickers and a shadow figure appears at the end of the hallway. The man freezes. The shadow moves unnaturally fast toward him. Loud whispering voices echo. The screen glitches, then cuts to black with a sudden scream. Ultra realistic, cinematic lighting, horror atmosphere, 4K, film grain, dramatic shadows, handheld camera, suspenseful tension.
【风格】高级商业氛围短片(Premium Lifestyle Commercial),高清手机前置镜头质感(Vlog互动视角),电影写实色彩,面部柔光打光,注意分镜编排和对口型,单人出镜 【时长】15秒 【场景】卧室+调酒区一体空间,温馨台灯与紫黄霓虹氛围光交织,微醺暖调 【角色】完全参考@ 的人物外貌特征(顶级亚洲明艳神颜),穿深蓝蕾丝吊带裙、长卷发,全程形象高度一致 [00:00-00:03] 镜头1:POV慵懒起床(POV → Medium Shot 丝滑转场) 主观视角(POV),镜头从第一人称俯视角度对准床上盖着毯子的双腿,手伸入画面掀开被子,双腿落地起身。镜头随起身动作自然抬起并快速切至中景——女主已站在床边面对镜头,慵懒又带一丝霸道地说:"睡什么睡,起来喝酒。"(中文口型精准同步) [00:03-00:06] 镜头2:优雅倒酒(Medium Shot) 中景,机位切至卧室调酒区,背景台灯暖光与紫黄霓虹光映在墙面。女主拿起白酒瓶,手腕微倾,透明酒液沿瓶口流入银色量酒器,随后将量酒器中的酒液倒入杯中并加入橙汁,液体混合的色彩层次清晰可见。 [00:06-00:09] 镜头3:挤柠檬特写(Close-up) 近景动作特写。女主双手各持半个切开的青柠,用力对挤,柠檬汁水真实地滴落进调酒壶,汁液飞溅的细节清晰,随后抓起一把冰块丢入壶中,冰块碰撞发出清脆声响。 [00:09-00:12] 镜头4:摇酒(Close-up / Shake) 近景。女主双手持调酒壶举至脸侧,有节奏地快速摇晃,长卷发随动作轻微摆动,壶内冰块撞击声清脆卡点。她保持迷人微笑注视镜头,眼神明亮带笑。 [00:12-00:15] 镜头5:品尝与收尾(Medium Shot) 中景。女主手持一杯蓝黄渐变的精致鸡尾酒,杯中气泡缓缓上升。她轻抿一口,眉眼舒展露出满意的微醺神情,随后对镜头开心挥手,定格。
Cinematic dark fairy-tale thriller with comedic twist, eerie enchanted forest, low-key lighting, cold moonlight filtering through dense trees, volumetric fog, high contrast shadows, dynamic handheld camera, expressive character animation, fast-paced storytelling. S1 (0.9s) Wide shot: An orange cat older brother in blue overalls and a gray flat cap runs barefoot through a dark forest, pulling along a white cat little sister in a dirty pink dress, both breathing heavily. S2 (0.8s) Rear tracking shot: The forest behind them echoes with distant howling wolves, shadows shifting between trees. S3 (0.9s) Forward reveal: Through the fog, a glowing candy house made of sweets appears at the end of the path, colorful and surreal against the dark woods. S4 (0.7s) Close-up: The siblings slow down, eyes wide—fear mixed with curiosity. S5 (0.8s) Door action: The orange cat cautiously pushes open the candy door, it creaks softly. S6 (0.9s) Interior wide: Inside, dim warm light reveals a strange room—giant half-open birdcage, scattered sweets, and a bed in the corner. S7 (0.8s) Close-up: On the bed lies a white-haired “grandma” in green pajamas and a white shower cap, blanket half covering her face. S8 (0.9s) Dialogue moment: The “grandma” weakly gestures toward the birdcage, implying there is cake nearby, voice soft and inviting. S9 (0.8s) Insert shot: A cake sits temptingly near the open birdcage door. S10 (1.0s) Action split: The little sister eagerly reaches for the cake, while the orange cat narrows his eyes, scanning the room. S11 (0.9s) Suspicion reveal: Close-up at the foot of the bed—a wolf tail sticks out from beneath the blanket. S12 (1.0s) Action burst: The orange cat instantly grabs a nearby large iron shovel, swings it with full force toward the “grandma’s” head. S13 (1.0s) Impact & reveal: The shower cap flies off—“grandma” leaps up in pain, revealing a ferocious wolf face, snarling before panicking. S14 (1.0s) Escape: The wolf stumbles and scrambles out the door, fleeing into the dark forest. S15 (1.2s) Final comedic beat: The orange cat scratches the back of his head proudly, while the little sister happily eats cake beside him, cheering with crumbs on her face. Motion & Effects: Dynamic chase camera, shaky handheld for tension, fast snap zooms during action, exaggerated character expressions, cloth and fur simulation, soft glow from candy elements, contrast between horror atmosphere and playful comedic timing.
15-second vertical emergency street incident, realistic cinematic documentary style, 9:16. Reference binding: @ Image1 is the main subject and must remain the same adult young woman throughout the full video. Preserve her exact face, long straight dark-brown hair with center part, fair skin, soft oval face, defined eyeliner, glossy pale pink lips, multiple silver ear studs, and yellow one-shoulder knit top with a silver oval brooch. Do not age-shift her. Do not change hairstyle, outfit, or identity. Scene: daytime in Brooklyn, New York, neighborhood sidewalk outside a corner deli or small storefront, brick wall, brownstone-like street character, parked cars along the curb, lived-in local block energy, sudden public danger, raw neighborhood realism. Core action: a large black bull with thick curved horns has @ Image1 pinned against the brick wall on the sidewalk. She is terrified, gripping both horns with both hands, struggling constantly to stop the bull from pressing into her. The bull pushes forward with believable force and jerks its head aggressively. Nearby locals gather fast, shouting, backing away, and trying to help. One man inches in from the side. Another person yells to call 911. Audio: realistic Brooklyn street ambience, passing cars, distant siren, footsteps, neighborhood voices, frightened screams, bull snorting, urgent English crowd shouting, no music, no narration. The woman cries: “Help! Please help!” Bystanders shout: “Call 911!” “Yo, back up!” “Somebody help her!” Shot timeline: 0.0-3.5s: medium-wide handheld shot from across the sidewalk. The bull has @ Image1 pinned to the wall outside the storefront. She grips both horns, face locked in panic, yelling for help. Nearby pedestrians stop abruptly. 3.5-7.0s: tight side-angle shot. The bull surges forward and tosses its head. She braces and struggles, elbows shaking, shoes dragging against the pavement as she fights to keep the horns away. 7.0-11.0s: wider vertical shot reveals more of the Brooklyn block. People gather near parked cars and the storefront, some filming, some retreating. One bystander steps in but pulls back when the bull snaps toward him. 11.0-15.0s: shaky close handheld climax. Her grip slips for a moment, then clamps down again. The bull snorts and presses harder. Two bystanders wave from the side to distract it while others scream for emergency help. End on unresolved danger. Camera: vertical bystander-phone footage, handheld, reactive reframing, slight panic zoom and imperfect centering, no impossible moves, no slow motion. Visual rules: realistic anatomy, realistic bull movement, realistic neighborhood crowd behavior, authentic Brooklyn street mood, no comedy, no fantasy, no gore, no blood spray, no severe visible injury, no subtitles, no text overlay, no watermark, no duplicated people.
women flying across the city like superwoman, holding a can of Coca-Cola in hand, wide-angle lens view from the front, close-up focus on the can, city in dynamic motion blur, cinematic lighting, ultra-realistic, high speed action --ar 3:4 --style raw --v 7
A dreamy cinematic landscape, a lone tree standing in the middle of a vast meadow filled with soft pink wildflowers. In the background, a gigantic towering white cumulus cloud rises dramatically into a deep blue sky. The foreground flowers are slightly out of focus, creating a beautiful shallow depth of field and bokeh effect. The camera slowly pushes forward through the flower field toward the tree, creating a peaceful cinematic feeling. Gentle wind moves the flowers and grass naturally. Soft sunlight illuminates the clouds, giving them a glowing, volumetric look. Ultra realistic nature cinematography, HDR lighting, soft color grading, dreamy atmosphere, cinematic depth of field, 4K, vertical composition, slow motion nature shot, highly detailed clouds, calm and peaceful mood.
A dramatic cinematic sequence featuring a fierce East Asian female cyborg warrior with short black bob hair in a dystopian futuristic city under a smoky, overcast sky. The video begins with a close-up on her sleek black robotic right hand, detailed metallic fingers with glowing red lights pulsing along the joints, subtle scratches and battle damage visible on the armor. The camera slowly tilts upward, revealing her heavily damaged yet powerful full-body tactical cybernetic armor — matte black with red glowing accents, exposed mechanical parts, and blood stains on her face. Her face shows intense emotion: furrowed brows, blood dripping from her mouth and cuts on her cheek, glowing red cybernetic eyes with horizontal red light beams scanning across them. She blinks slowly, her expression shifting from pain to determination. She gradually stands taller in the ruined urban battlefield filled with smoke, distant fires, and destroyed buildings. She puts on a sleek black tactical helmet with a dark visor. The camera pulls back to a full-body shot as she turns her body confidently, revealing the full armor design with shoulder pads, thigh holsters, and back-mounted weapons. Red energy lines pulse across her suit. In the final moments, she activates her thrusters — bright orange-white flames burst from her hands and feet. She launches powerfully into the air, ascending rapidly above the destroyed cityscape as explosions and thick black smoke trails fill the background. Epic, intense, cyberpunk atmosphere, highly detailed, cinematic lighting, slow-motion elements mixed with dynamic movement, dark moody color grading with strong contrast on the red glows. cinematic, ultra realistic, 8k, photorealistic, dramatic lighting, film grain, intense action sequence, 12 seconds, smooth camera movement, slow tilt up then pull back, final rocket launch
A super cute chibi-style anime cartoon little penguin girl with big sparkling eyes, fluffy black and white feathers, tiny wings, and a small pink bow on her head. She's sitting at a table happily eating her favorite food with chopsticks. Suddenly, a finger gently pokes her soft cheek from the side. Show her dramatic adorable reaction in sequence: eyes widening in surprise, cheeks puffing out, intense blush, a funny puffed-up angry-cute expression with tiny steam coming from her head, then she turns and glares cutely at the finger while still chewing. Kawaii anime style, soft pastel colors, smooth 2D animation, vibrant and wholesome, high detail, dynamic short animation sequence, funny and heartwarming mood, clean bright background.
Extreme fractal zoom (Mandelbrot effect) centered on the eye of a person standing in Shibuya. The camera zooms infinitely into his pupil, revealing another Shibuya crossing inside, over and over at high speed. The buildings and people stretch and twist into kaleidoscopic geometric patterns. Mind-bending recursive VFX, psychedelic motion, infinite perspective.
[CINEMATIC SETUP]Film Stock: 35mm gritty film grain, handheld documentary-style movement.Lenses: 24mm wide-angle for the chase, switching to a high-speed 85mm prime for the finale.Color Grade: Raw, gritty, desaturated jungle greens with high-contrast shadows.Lighting: Dappled natural sunlight breaking through thick canopy, dramatic volumetric god rays.Audio Style: Immersive spatial sound design; heavy breathing, splashing water, the high-pitched “thwp” of blowgun darts, and a deep sub-bass impact during the catch.[IMAGE REFERENCES / LEGEND] @Image 1 : The exact main character; keep the bearded face, white technical rain jacket, and black cap exactly as seen in the reference.[TIMELINE SECOND BY SECOND] 0-4s: [Low-angle handheld tracking shot] @Image 1 runs desperately toward the camera through dense jungle foliage. Blurred indigenous figures are visible in the background shadows. Sound: Panic-filled heavy breathing and snapping branches.4-7s: [Cinematic side-profile tracking] @Image 1 sprints past a massive ancient tree. A primitive wooden dart whistles through the air and embeds itself into the bark with a sharp “thwack” just inches behind his head.7-11s: [Wide shot to Medium] @Image 1 reaches the edge of a massive waterfall precipice. He teeters on the edge, arms flailing as he almost loses his balance over the mist; he then lunges forward, digging his heels in to successfully regain his footing.11-15s: [Super slow-motion / Bullet time] @Image1 snaps his head toward a new dart approaching at high speed. The camera circles him as he reaches out in ultra-slow motion to snatch the dart mid-air. Final 0.5s: The motion suddenly ultra-accelerates to real-time as he slams the dart into the ground and holds it firm. Sound: Sound fades to a heartbeat, ending with a sharp metallic “crack” impact upon the catch.[STYLE & QUALITY BOOSTERS] Photorealistic 8K, ultra-detailed fabric textures on the white jacket, movie-level realistic facial features, no deformation, stable motion throughout, high dynamic range.
aesthetic: phone held up in a packed sports bar, wall of TVs glowing behind audio: crowd chant building, stomping getting louder, bass underneath timeline: - "0-4s: Phone camera inside a packed sports bar. Hundreds of people. Wall of screens behind them. Standing ON the bar counter is an 81-year-old man in a cardigan, khakis, and loafers. A beer in each hand. The crowd is chanting." - "4-8s: The drop hits. He slams both beers, crushes the cups, and starts stomping on the bar. Glasses and bottles jump. The bartender ducks. The crowd goes berserk." - "8-15s: He's dancing on the bar, loafers sliding on the wet surface. People reaching up to high-five him. He points at someone in the crowd and flexes. The cardigan is coming apart. He does not care." quality: single shot from below the bar looking up, TV glow, the old man towering over the crowd, beer foam everywhere
Deep within the subterranean tunnels, a gargantuan technician meticulously rewires a pulsing energy core. Each spark and hum tells a story of a civilization powered by forgotten tech and raw mechanical strength. A fusion of organic mass and heavy metal The hypnotic glow of high-voltage circuits Experience the electrical surge... (with sound)