Cityscape / Urban

Ink wash shrimp leaping into a futuristic city—a piece of rice paper unfolds among nocturnal skyscrapers, eight black-and-white ink wash shrimp leap out from the paper's surface, their bodies still retaining the 'flying white' brushwork and ink gradients. The shrimp whiskers turn into glowing city tracks, and ink droplets become neon rain; the scene features huge scale contrast, strong depth, cinematic lighting, and a spectacle that captures the eye instantly. No text, no watermarks.

Cinematic cyberpunk alley at night during heavy rain, neon signs reflecting on wet pavement in pink, cyan and violet. A stunning Japanese woman in her mid-20s with long black hair, soft face, wearing a modern cyber-geisha outfit: elegant black silk kimono with glowing neon embroidery, high slit, thigh-high boots. She is gently cradling a fluffy white and silver cat with big luminous eyes and tiny glowing cybernetic ear implants. Soft rain droplets on their faces and fur. Shallow depth of field, volumetric god rays from neon signs, ultra-detailed, emotional and serene atmosphere, shot on 85mm lens, cinematic lighting, high contrast, 8k

Quiet suburban street, evening, sprinklers running. Sky splits with a silent lightning-like tear, widening into a churning purple void. Camera pushes toward it from ground level as debris starts lifting off the street; ends with the camera crossing the threshold, the suburb replaced instantly by an alien horizon.

[Overall Style] Deadpan supernatural reversal comedy, opening with real cinematic texture, disguised as an oppressive urban ghost story in the first half, revealing it as a common rainy night misunderstanding in the second half. Featuring cold cyan streetlights, warm light from a convenience store, reflections on wet asphalt, restrained facial performances, precise spatial staging, and a rhythmic transition from horror sound effects to comedic audio. [Characters] Character A | Sword Immortal Sister Use @Image 1 as strict identity and costume reference: same 25-30 year old East Asian female, oval face, sharp dark eyes, long straight black hair, fair natural skin, tall slender figure, wearing white embroidered silk Hanfu, translucent layered wide sleeves, silver waist ornaments, jade hairpin, and white cloth boots. Character B | Bike Sister Use @Image 2 as strict reference: same 25-30 year old East Asian female, maintaining the same face, ponytail, body proportions, clothing, accessories, and the same bicycle covered by a translucent white raincoat. [Shot 1 | 0-5s | Low Angle Wide Slow Tracking] 16:9, empty modern urban underground passage at night, heavy rain, concrete pillars, dripping pipes, flickering fluorescent lights, distant neon signs, scattered leaves, and heavy mist clearly visible. Character A stands fully in the light, holding the same silver sword. Character B silently approaches on the bike covered in the raincoat; the reflectors look like glowing eyes, a plastic bag flapping like a ghost's tongue. Low horror ambient sound, rain, and bike chain noise. [Shot 2 | 5-10s | Medium Shot] Character A suddenly turns, draws the sword, sleeves billowing, shouting: "Show yourself, monster!" The raincoat figure doesn't answer, gliding forward steadily. Camera tracks back at the same speed, horror frequency intensifies, then a small crisp bike bell rings. [Shot 3 | 10-15s | Close-up to Two-Shot] Character B brakes under convenience store light, lifts the raincoat hood, revealing the consistent face, saying expressionlessly: "Sister, I just got off work. You're the one who looks like a ghost." Heroic sword sound stops instantly. Extreme close-up: Character A looks at her white Hanfu, hides the sword behind her back, nods while trying to remain cool, one eyelid twitching slightly. Freeze on Character B staring at her. [Technical Requirements] Strict 15s duration, three shots, stable character identity, coherent motion direction, realistic physics for rain, hair, and fabric, native Mandarin dialogue and spatial audio, no subtitles. Optimized for Seedance 2.0 multimodal input. [Negative Prompts] blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, face morphing, background shift, glitching cuts, disappearing props

A stylish young woman walks through a crowded city street during golden hour. With a simple snap of her fingers, time freezes instantly—cars stop mid-motion, birds remain suspended in the sky, raindrops hang in the air, and every person becomes perfectly still. She calmly explores the silent world, walking between frozen moments with cinematic tracking shots and breathtaking visuals. She then notices a little girl moments away from being struck by a falling construction sign. While time stands still, she gently moves the child to safety. With one final snap, time resumes. The sign crashes harmlessly to the ground, the crowd continues as if nothing happened, and the mysterious woman quietly walks away with a smile. Ultra-realistic visuals, emotional storytelling, golden-hour lighting, seamless transitions, and a powerful cinematic ending.

Camera flying fast and low through cloud layer, then bursts through into a floating city — spires, hanging gardens, glowing transit rails looping between towers. Drone-style continuous forward push, banking hard around a tower, market stalls and citizens glancing up, ending as camera dives off the city's edge into open sky below, revealing the ocean thousands of feet down.

Main Subject: Young Korean woman in her early 20s with a natural everyday appearance, realistic skin texture, minimal makeup, expressive eyes, warm and approachable personality. She wears a faded charcoal-grey sleeveless crop top, loose high-waisted light-wash jeans, black canvas sneakers, and a simple black cord necklace. Her black wavy hair is tied in a messy side ponytail with soft wispy bangs. Maintain identical facial identity, hairstyle, clothing, proportions, and accessories throughout the entire video. Location: A quiet Japanese-style suburban neighborhood during golden hour, lined with cherry trees, small traditional homes, narrow streets, overhead power lines, blooming flowers, and a peaceful stream crossed by a small wooden bridge. Soft summer breeze, birds chirping, and warm sunlight create a calm nostalgic atmosphere. 00:00–00:03 She walks outside her house, smiles softly, unlocks an old vintage bicycle, and gently pushes it onto the quiet street. Natural body language, subtle breathing, realistic hand interaction, handheld documentary-style camera. 00:03–00:06 She begins riding slowly through the peaceful neighborhood. The camera lightly jogs behind her with gentle handheld movement. Her ponytail sways naturally in the breeze as sunlight filters through the trees. 00:06–00:09 She happily rings the bicycle bell while riding past blooming flowers and tree-lined streets. Her smile feels genuine and effortless, with realistic micro-expressions, blinking, and natural eye movement. 00:09–00:12 She slows down and stops beside a small wooden bridge overlooking a narrow flowing stream. The camera moves around her as she quietly enjoys the peaceful scenery while leaves gently move in the wind. 00:12–00:15 She looks back toward the camera with a warm nostalgic smile before riding away into the golden sunset. The camera lingers as she disappears down the quiet street, leaving a calm, emotionally memorable ending. Visual Style: Ultra-photorealistic live-action, 8K HDR, cinematic handheld documentary camera, natural micro-shakes, Kodak Vision3 35mm film look, soft film grain, warm golden-hour lighting, physically accurate global illumination, volumetric sunlight, realistic skin pores, natural hair simulation, accurate cloth physics, premium cinematic colour grading, shallow depth of field, subtle lens breathing, emotionally immersive slice-of-life storytelling, every frame indistinguishable from a high-budget feature film.

Duration: 15 seconds Aspect Ratio: 16:9 Style: Ultra-photorealistic kaiju disaster film with realistic structural destruction, stormy coastal metropolis, physically accurate debris simulation, cinematic action, IMAX-scale realism. Monster Design A colossal deep-sea octopus towering above nearby skyscrapers. Its wet crimson-purple skin is scarred and textured, with pale lifeless eyes, rough oversized suction cups, a battle-worn mantle, and eight enormous muscular tentacles. Movement: The creature never stops moving or poses. Four tentacles continuously pull its massive body forward at high speed while the remaining four violently destroy everything in its path. All eight tentacles remain fully connected, clearly separated, anatomically correct, and constantly visible throughout the sequence. 0:00–0:02 | Street Assault Camera: Low street-level tracking shot, smoothly pulling backward. Action: Destruction begins immediately. The monster charges directly toward the camera. One tentacle crushes a city bus and several streetlights while another wraps tightly around a glass office tower. The remaining tentacles drag the creature rapidly through flooded streets. Lighting: Cold storm daylight with flashing red emergency lights reflecting across wet pavement. VFX: Heavy rain, splashing water, shattered glass, broken asphalt, flying debris. 0:02–0:05 | Tower Collapse Camera: Wide rooftop shot with smooth lateral tracking. Action: Without slowing down, the octopus violently yanks the glass tower sideways. Lower floors rip apart under extreme stress before the structure collapses behind the advancing creature. Lighting: Lightning flashes illuminate dark storm clouds. VFX: Concrete dust, electrical sparks, collapsing steel beams, cascading glass. 0:05–0:08 | City Charge Camera: High aerial tracking shot with controlled movement. Action: The monster surges through multiple intersections at incredible speed. Every tentacle strike fractures the pavement, launches vehicles through the air, and tears apart surrounding infrastructure while its body never loses momentum. Lighting: Steel-blue rainstorm daylight. VFX: Exploding puddles, cracked roads, smoke, debris trails. 0:08–0:12 | Block Leap Camera: Wide frontal shot, smoothly retreating. Action: The creature plants two massive tentacles onto buildings on opposite sides of the street, compresses its enormous body, then launches across an entire city block. During the leap its remaining tentacles rip away balconies, billboards, rooftop equipment, and facades while maintaining continuous forward motion. Lighting: Dramatic storm backlight mixed with orange firelight. VFX: Flying debris trails, displaced rain, drifting smoke, shattered concrete. 0:12–0:15 | Final Charge Camera: Ultra-wide cinematic shot slowly pulling back. Action: The monster lands near the coastline

Create an ultra-realistic cinematic short film** featuring the same young East Asian woman consistently throughout the entire film. Maintain identical facial features, hairstyle, body proportions, and outfit continuity. She wears a fitted white crop top, light blue high-waisted jeans, white sneakers, a delicate silver necklace, and carries a small cream crossbody bag. The mysterious young man wears a beige overshirt, white T-shirt, blue jeans, and white sneakers. Premium Korean slice-of-life aesthetic, warm natural lighting, realistic performances, elegant handheld and gimbal camera movement, shallow depth of field, soft cinematic color grading, photorealistic 4K HDR, 16:9 widescreen. The film opens on a peaceful Saturday morning. The young woman decides to spend the day exploring her city with a vintage disposable camera she recently found at home. She wanders through colorful streets, flower markets, quiet cafés, bookstores, and riverside paths, capturing small moments that make her smile. She photographs blooming flowers, street musicians, playful children chasing pigeons, fresh pastries in a bakery window, and her reflection in a café mirror. Every click of the disposable camera captures another memory. As the afternoon passes, she sits alone on a park bench overlooking a quiet lake. She finishes her coffee, smiles while looking through the camera one last time, then receives a phone call. In a hurry, she stands up and walks away, accidentally leaving the disposable camera behind on the bench. A few minutes later, a kind young man walking through the park notices the forgotten camera. He looks around but cannot find its owner. Curious, he gently picks it up and notices there is only one exposure remaining. Instead of keeping it, he smiles and takes a single photograph of himself holding a handwritten note that says, **"Your memories deserved to come home."** He carefully develops the entire roll of film at a local photo lab. The printed photographs beautifully capture the woman's joyful day. On the back of the final photograph, he writes the address of the café where he found the camera and leaves the developed photos inside a neatly wrapped envelope with the camera. A few days later, the woman receives the package after the photo lab traces her contact information from the camera purchase receipt. Sitting in her bedroom, she slowly flips through every printed photograph, smiling as forgotten moments come back to life. When she reaches the final photo, she sees the stranger holding the handwritten note. Curious and touched by the gesture, she notices the café address written on the back. The following afternoon, she visits the small café. As she walks inside holding the disposable camera, she immediately recognizes the young man sitting near the window reading a book. Their eyes meet, and they exchange warm, slightly nervous smi

[Overall Setting] Generated Video: A deadpan meta-comedy with a realistic cinematic texture. It begins disguised as a high-budget Xianxia visual spectacle and ends by revealing the 'magic' is actually the hard labor of an ordinary person. The acting is restrained, visual misdirection is clear, spatial relations are coherent, and water physics are realistic, concluded by a reveal through purposeful camera movement. [Character References] Use @Image 1 as the strict identity and clothing reference for Character A: The same 25-30 year old East Asian female sword immortal, oval face, sharp dark eyes, long black hair, tall and slender build, wearing white embroidered silk Hanfu, translucent layered wide sleeves, silver waist ornaments, jade hairpin, and white cloth boots. Use @Image 2 as the strict identity and clothing reference for Character B: The same 25-30 year old East Asian cycling girl, strictly maintaining the same face, hairstyle, body proportions, clothing, accessories, and bicycle as in the reference photo. [Shot 1 | 0-3s | Low-angle wide shot] A modern riverside plaza bright with sunlight after rain. The wet stone ground creates clear reflections, surrounded by glass buildings, dripping trees, and light mist. The same bicycle is parked next to a low concrete planter. The same sword immortal enters the frame completely from feet to head, holding the same silver longsword. A shimmering stream of water rises behind her, spiraling around the bicycle to form a grand, transparent water dragon. The camera slowly and steadily pushes in, accompanied by heroic orchestral music, the whistling of water, and the sound of fluttering silk. [Shot 2 | 3-7s | Cowboy medium shot] The same sword immortal in white silk Hanfu spins her sword, guiding the same water dragon in an elegant spiral. Water droplets glisten in backlighting, her long hair and wide sleeves flutter dramatically. She says calmly and majestically: 'This is the Water Control Art.' The camera begins a continuous, stable semi-circular pan around her. The background consistently remains the same plaza, planter, bicycle, glass buildings, and wet ground. [Shot 3 | 7-10s | Reveal close-up] The camera completes its semi-circular movement, revealing the same cycling girl (@Image 2) squatting tiredly behind the planter, one hand holding a high-pressure water gun and the other holding a portable blower. The water hose suddenly kinks; the grand water dragon instantly loses form and drenches the sword immortal's head, flattening her hair and soaking her white Hanfu. The heroic music cuts off abruptly. The cycling girl puts down the blower, rubs her aching arm, and says expressionlessly: 'Sis, can the magic stop now? My arm is sore.' An extreme close-up of the sword immortal's dripping eyelashes and frozen, sideways stare. [Technical Requirements] Strict total duration of 10 seconds, three shots, 16:9 aspect ratio, stable character identities, accurate Mandarin lip-sync, precise synchronization of water flow, water gun, blower, and music. Do not generate on-screen subtitles. [Negative Prompts] blurry, bad quality, low quality, low resolution, noisy, jpeg artifacts, watermark, text, error; deformed, mutated, bad anatomy, poorly drawn hands, bad composition, out of frame, disfigured; inconsistent character, changing clothes, face morphing, background shift, glitching cuts, disappearing props

Use the uploaded character reference sheet as the ONLY identity reference. Preserve the exact facial identity, hairstyle, beard, facial proportions, skin tone, body build, and clothing from the reference sheet throughout the entire video. Never replace or alter the character's face. Keep the appearance perfectly consistent in every frame. A 12-second ultra-realistic cinematic superhero action sequence on a flooded, rain-soaked New York City street at night during a heavy thunderstorm. Tall skyscrapers tower on both sides, neon signs and digital billboards glow through the rain, wet asphalt and puddles reflect lightning flashes, streetlights, and distant city lights. Heavy rain pours relentlessly, creating thick mist and dramatic storm clouds overhead. The camera performs one uninterrupted continuous shot with seamless movement. The reference character stands in the middle of the empty street in a calm fighting stance as Spider-Man lands several meters away after swinging down from a nearby building. Rain splashes realistically from every movement. The camera slowly circles both fighters while pushing in, creating intense cinematic tension. Spider-Man suddenly charges forward at superhuman speed, throwing a rapid combination of punches and spinning kicks. The reference character expertly blocks, dodges, and counters with realistic martial arts techniques. Every impact creates cinematic slow-motion rain droplets, flying water particles, realistic cloth movement, and subtle camera shake. The fight feels grounded yet highly athletic. Spider-Man leaps into the air attempting an aerial kick. The reference character narrowly avoids the attack, grabs Spider-Man's arm, and redirects his momentum across the wet street. Spider-Man immediately fires web lines toward nearby street lamps and building facades, swings around them, and launches back into combat. Sparks fly as both fighters slide across the soaked asphalt and metal surfaces while rain pours heavily. The camera continuously follows the action with dynamic orbiting movements, low-angle hero shots, close-up facial reactions, and wide cinematic reveals of the glowing city skyline. Lightning flashes dramatically behind both characters, creating high-contrast silhouettes and realistic reflections on the wet street. The battle reaches its climax as both fighters sprint directly toward each other. The reference character throws a powerful punch while Spider-Man delivers a flying kick. Their attacks collide at the exact same moment, producing a cinematic shockwave of rain and debris. Time briefly slows as lightning strikes

Modern Rainy Square × Xianxia Deadpan Reversal Comedy [Style] Cinematic realistic deadpan comedy, presenting elegant Xianxia wonders clashing with uncontrollable daily weather. Optimized for Seedance 2.0 multimodal control and native audio-sync. [Characters] Character A: Sword Immortal Sister from @Image 1. 25-30 year old East Asian female. Strict identity/clothes locking (white silk Hanfu, jade hairpin, etc.). Personality: calm, proud, restrained. Character B: Bike Sister from @Image 2. 25-30 year old East Asian female. Yellow jacket, jeans, sneakers, same bike. Personality: natural, practical. [Shot 1 | 0-3s] Wide shot. Rainy urban square, wet stone reflections. Character B stops at a bike rack/umbrella stand. A red umbrella blows away. Character A enters, calmly raising two fingers to observe. Sound: rain, distant traffic, subtle sword hum. [Shot 2 | 3-6s] Medium shot, controlled orbit. Character A rotates her wrist. Eight umbrellas rise from the stand and open into a symmetrical circular array like a magical formation. She says: "Small scene." Camera orbits to show the structure. [Shot 3 | 6-10s] A sudden strong wind fills the umbrellas, turning them into a giant parachute. Character A is pulled vertically off the ground. Her expression shifts from proud to silent panic. Character B picks up the red umbrella, opens it normally, and says: "Sister, are you holding an umbrella or taking off?" Close-up on Character A's twitching eyes while she tries to maintain her pose. Freeze on the flight. [Action Sequence Lock] 1. Wind intensifies. 2. Umbrellas fill. 3. Becomes parachute. 4. Character A is pulled up. 5. B picks up red umbrella. 6. B delivers line. 7. Close-up on A's face. 8. Freeze. [Technical Requirements] Strict 10s, three shots, realistic physics for wind/fabric/weight, Mandarin lip-sync, no subtitles, no body duplication or background drift.

Build the monster, city, scale, and destruction entirely from text." style: "Photorealistic kaiju disaster film, stormy coastal metropolis, grounded structural destruction." visual_feel: "Rapid creature movement, controlled camera, heavy impacts, rain, spray, dust, readable tentacle action." duration: "15 seconds" character_modeling: octopus_monster: base: "A colossal deep-sea octopus taller than nearby skyscrapers." features: "Dark crimson-purple wet skin, pale eyes, rough suction cups, scarred mantle, eight massive muscular tentacles." movement_in_this_scene: "It never stops or poses. Four tentacles pull its body rapidly forward while the others destroy obstacles." cinematic_storyboard: 00_02_street_attack: camera: "Low street tracking backward, steady." action: "The monster crashes in immediately. One tentacle crushes buses and streetlights while another grips a glass tower." lighting: "Cold storm daylight, red siren reflections." vfx: "Water spray, glass, asphalt debris." 02_05_tower_pull: camera: "Wide rooftop side tracking." action: "Still advancing, it yanks the tower sideways. Lower floors shear apart and the structure collapses behind it." lighting: "Lightning through dark clouds." vfx: "Concrete dust, sparks, falling glass." 05_08_city_run: camera: "High aerial follow, controlled." action: "It races through several intersections. Each tentacle strike cracks roads and throws vehicles aside." lighting: "Steel-blue rain light." vfx: "Puddle bursts, fractured pavement, smoke." 02_12_block_vault: camera: "Wide frontal pull-back." action: "The monster plants two tentacles on opposite buildings, compresses, then vaults across a full city block, tearing signs and balconies free." lighting: "Storm backlight and orange fire glow." vfx: "Debris wake, rain trails, displaced smoke." 12_15_final_charge: camera: "Extreme wide slow pull-out." action: "It lands near the waterfront, sweeps through cranes and warehouses, then surges toward downtown as towers collapse behind it." lighting: "Lightning silhouette over burning streets." vfx: "Water plume, collapsing cranes, dust waves, embers." production_notes: audio_design: "Deep creature roars, rain, sirens, metal tearing, heavy impact bass." critical_constraint: "Destruction begins on frame one. Keep all eight tentacles connected, distinct, and readable. Fast creature action with controlled camera." avoid: "Slow opening, idle posing, extra limbs, tangled tentacles, anatomy drift, temporal flicker, weightless debris, chaotic camera."

Duration: 60 Seconds Aspect Ratio: 16:9 Style: Ultra-realistic cinematic travel film, 4K HDR, IMAX-quality visuals, premium color grading, fast-paced editing, whip transitions, speed ramps, seamless match cuts, energetic music, no dialogue. Concept: Experience one unforgettable day in Japan through a fast-paced cinematic journey that feels like a movie rather than a vlog. Sequence 0–5s | Hook Shibuya Crossing Tokyo Tower Neon streets Bullet train speeding past 5–20s | Tokyo Life Vending machine Convenience store breakfast Arcade games Anime stores Gashapon capsules Sushi and ramen 20–40s | Traditional Japan Kyoto temples Red torii gates Bamboo forest Matcha tea ceremony Cherry blossoms 40–55s | Adventure Mount Fuji Bullet train Nara deer Scenic countryside 55–60s | Finale Rainy neon streets Tokyo skyline at night Drone pullback over the city Ending Text: "See you again, Japan." 🇯🇵

Create a highly realistic, cinematic 15-second video featuring the same female character throughout every scene. She is a 24-year-old woman with long dark brown hair in a neat ponytail, expressive brown eyes, light olive skin, and a stylish casual outfit consisting of a black leather jacket, white oversized T-shirt, blue jeans, and white sneakers. Maintain identical facial features, hairstyle, clothing, body proportions, and accessories across the entire video. Ultra-photorealistic, natural lighting, realistic human movement, Hollywood-quality cinematography, 4K HDR, 24fps. Scene 1 (0–3s):The woman is walking through a busy modern city street when a glowing futuristic smartwatch on her wrist suddenly activates and displays "INVISIBLE – 00:15". Her body fades completely from view, leaving only the environment visible. Nearby pedestrians continue walking, completely unaware that she is there. Handheld cinematic camera, realistic city ambience. Scene 2 (3–6s):Still invisible, she playfully surprises people by gently moving a shopping cart on its own, making a café chair slide slightly, and causing a takeaway coffee cup to appear as if it is floating. The reactions are confused but humorous rather than frightening. Keep everything grounded and believable. Scene 3 (6–10s): She confidently walks through the entrance of an exclusive rooftop VIP event while security guards unknowingly let her pass. She casually picks up a small dessert and a sparkling drink from the buffet while invisible, smiling mischievously. Luxurious rooftop setting overlooking a city skyline at sunset. Scene 4 (10–15s):The smartwatch countdown rapidly reaches 00:03…00:02…00:01. She suddenly becomes fully visible again while standing in the middle of the VIP lounge holding the dessert and drink. The surrounding guests and security all turn toward her with shocked expressions. She freezes, gives an awkward smile, slowly raises one hand in an embarrassed wave, and the video ends on a comedic freeze frame. Natural crowd reactions, realistic facial expressions, cinematic ending. Style: Photorealistic, seamless continuity, smooth handheld camera movement, premium film-grade lighting, realistic physics, subtle humour, no subtitles, no watermarks, no visual glitches, no character changes, no extra limbs or distorted hands. The same woman must remain perfectly consistent in appearance throughout the entire video.

Main Subject: Young Korean woman, 23, natural everyday appearance, shoulder-length black hair worn loose beneath a cream knitted beanie, soft wispy bangs, realistic skin texture with rosy cheeks from the cold, minimal makeup. Wearing a beige wool coat over a chunky white knit sweater, dark straight-leg jeans, brown leather ankle boots, cream wool scarf, and black leather gloves. Maintain consistent identity, clothing, hairstyle, and appearance throughout. Location: Quiet residential street in Seoul during the season's first snowfall. Snow lightly covers rooftops, parked cars, narrow sidewalks, bicycles, utility poles, and bare trees. Warm light glows from apartment windows, creating a peaceful winter atmosphere. Visual Style: Ultra-realistic documentary realism. Genuine candid behaviour, natural body language, subtle expressions, authentic winter atmosphere, rich environmental details. Camera Style: Early-2000s Sony MiniDV camcorder. Heavy handheld shake, imperfect framing, autofocus hunting between falling snow and her face, exposure shifts from bright snow, faded colours, soft contrast, slight motion blur, MiniDV compression artifacts, light sensor noise, no stabilization. 00:00–00:03 She steps out of her apartment building, pauses, and smiles as snowflakes land on her scarf. The camera briefly focuses on the falling snow before finding her face. 00:03–00:06 Walking slowly down the snowy street, she leaves fresh footprints while brushing snow from a nearby railing. The operator struggles to keep her centred. 00:06–00:09 She stops beneath a bare tree, catches a few snowflakes on her black glove, then quietly laughs as one lands on her nose. 00:09–00:12 She buys a warm roasted sweet potato from a small street cart, holding it carefully in both hands as steam rises into the cold air. Autofocus shifts between the steam and her face. 00:12–00:15 She looks toward the camera with a warm smile, waves gently, and continues walking into the softly falling snow. The camera tilts slightly toward the snowy pavement before the recording abruptly cuts to black. Audio: Natural ambience only—light wind, soft footsteps in snow, distant traffic, birds, quiet neighbourhood chatter, the faint sizzle from the street cart, fabric rustling, and falling snow. No music. No narration. Goal: A forgotten MiniDV home video from winter 2004, capturing the quiet excitement of the first snowfall in an ordinary Seoul neighbourhood—warm, imperfect, nostalgic, and completely believable.

Create an ultra-realistic 15-second vertical (9:16) smartphone vlog. A young woman is filming herself with the front camera while walking through a quiet Tokyo street during golden hour. The camera has natural handheld movement, slight autofocus shifts, realistic lighting, and subtle background blur. People walk by naturally, bicycles pass in the background, birds chirp, and distant traffic creates authentic city ambience. She smiles warmly at the camera and speaks naturally in English. She walks into a cozy café, briefly flips the camera to show the coffee machine and barista making an iced latte, then switches back to selfie mode. She picks up the drink, walks outside, takes a sip, and continues walking with the city behind her. The video should feel like a genuine influencer vlog, not scripted or cinematic. Voice & Timing: 0–4 sec: "Good morning, everyone! I found the cutest little coffee shop while exploring Tokyo." 4–9 sec: "Let's see if their iced latte is really as good as everyone says." 9–15 sec: (Takes a sip and smiles) "Wow... that's amazing. Definitely worth the stop! See you in the next adventure!" Style & Audio: Ultra-photorealistic, 4K HDR Natural facial expressions and lip sync Realistic smartphone front-camera quality Ambient city sounds only (birds, footsteps, distant traffic, café sounds) No background music Authentic influencer vlog aesthetic Smooth transitions, realistic motion blur, and no AI-looking artifacts

Create a gritty cinematic action shot, 7 seconds, 16:9. Use dynamic action-movie cutting. Main subjects: @jin, a lean man in a worn knee-length dark coat over dark clothes, expressionless, fast and economical. @mishima, a young man in a black suit, holding a compact Glock 19 pistol with a mounted underbarrel tactical light. Setting: A Neo-East slum alley at blue-hour twilight — deep-blue dusk sky overhead, faint neon city towers far beyond, retro filament-style LED bulbs glowing amber as the dominant local light: scrap-metal walls, tangled cables overhead, wet reflective ground, humid haze. A makeshift food stall with a steaming pot, a hanging amber lantern and a tarp canopy stands close behind @mishima. The shot opens over @jin's shoulder from directly behind him — @jin's back stays to the camera the whole time and he never turns toward the lens — looking down the alley at @mishima as @jin drives forward into the charge. As he closes the gap, cut to a tighter side angle: @jin catches @mishima's gun wrist on the outside line and wrenches it upward, and in the same motion drives an elbow into @mishima's face. Cut to @mishima's head snapping back as he is hurled backward and SLAMS into the food stall behind him — [VFX: a wooden support post snaps, the tarp canopy collapses, the metal cooking pot flips into the air and its broth sprays outward, utensils scatter, and a gout of flame flares up from the toppled stove just beside him as the amber lantern shatters]. The shot resolves with @jin already low and rooted in the wreckage of the stall, gaze flat, no pause. Camera behavior: Over-the-shoulder from directly behind @jin with his back to camera throughout, steadicam tracking in with the charge, a hard cut on the elbow impact, framing fixed with no reversal. Real-time speed with the impact landing heavy. Global style: Gritty cyber-noir at blue-hour twilight, cinematic low-key amber light from the filament bulbs as the key with a cool deep-blue ambient fill, wet reflective surfaces, humid haze, cold and quiet tone, high texture detail, 35mm film grain. Audio: Rapid wet footsteps closing in, cloth friction, the wrench of a seized joint, a sharp elbow impact, a wooden post cracking, the metal pot clanging and broth splashing, a flame whooshing up, and the lantern shattering. No dialogue. Keep @jin and @mishima visually distinct, and use only these two referenced characters.