REFERENCE RULES Use @image2 as the opening-frame reference. Use @image4 @image5 @image6 and @image7 as identity, wardrobe, location and composition anchors for the beauty room, photography studio, runway and outdoor advertising shoot. Use @image9 as the required final-frame reference. The same fictional 27-year-old Western supermodel must remain recognizable throughout the entire video. Preserve her exact face, gray-green eyes, high cheekbones, light olive skin, dark-blonde hair, height and slender athletic proportions. She must never resemble a real celebrity or public figure. DAY-IN-THE-LIFE ENERGY Create a fast, elegant and relentless day-in-the-life montage: wake up and wash -> stride through the city -> makeup preparation -> studio photoshoot -> rush to the next venue -> runway show -> outdoor campaign shoot -> ride home -> wash and sleep. The dominant visual mechanism is a chain of reflections, flashes and physical wipes. Each transition must be caused by an existing element: bedding -> bathroom towel -> mirror reflection -> passing vehicle -> vanity light -> camera flash -> coat crossing the lens -> car door -> runway curtain -> flashbulb -> sunlight flare -> car-window reflection -> bathroom water -> bedside darkness. Do not use arbitrary dissolves or magical transformations. WARDROBE STATE RULE Wardrobe changes occur only while the frame is fully obscured by a flash, door, curtain, coat or mirror transition. Never morph clothing visibly on the body. Wardrobe sequence: charcoal-gray pajamas -> camel street coat with black turtleneck and tailored black trousers -> black salon robe -> crimson couture studio dress -> black travel coat -> black sculptural runway gown -> ivory trench with silver-gray tailored outfit -> the same ivory trench inside the return car -> charcoal-gray pajamas at home. SHOT 1 — WAKE, WATER AND IMMEDIATE DEPARTURE Begin exactly from @image2 . A smartphone vibrates on the bedside table. Her eyes open instantly and her hand stops the vibration. The camera snaps into a rapid push toward her face as she throws the duvet upward. The white bedding fills the entire frame and becomes a motivated wipe into the bathroom. As the fabric clears, she is already at the sink in charcoal-gray pajamas. Use the mirror as the main perspective. She splashes cold water onto her face, runs both hands backward through her hair and looks up sharply into her reflection. Water droplets strike the mirror and briefly blur the lens. She pulls a white towel across the mirror. When it clears, the bathroom reflection has become a bright storefront reflection on the city street. Keep the awakening and washing concise and physically readable. SHOT 2 — STREET POWER WALK INTO THE VANITY The same model strides quickly along the city pavement in the camel oversized coat, black turtleneck, tailored black trousers, boots, sunglasses and leather tote. Use a low lateral tracking shot close to her boots, then accelerate into a backward-facing medium shot as she crosses the street. Pedestrians and cyclists move around her at normal speed while she maintains a clean, confident line through the crowd. Her coat, hair and tote respond naturally to her pace. Storefront reflections repeatedly reveal her face from different angles without duplicating her. A black car passes extremely close across the foreground, completely covering the frame. Its dark reflective surface transforms naturally into the black edge of the illuminated vanity mirror. SHOT 3 — RAPID BEAUTY PREPARATION Reveal @image4 . Use a tight orbit around the vanity rather than a static mirror shot. A makeup brush sweeps across her cheek. A hair stylist tightens the sleek low ponytail. Lip color is applied in one precise motion. She remains still only long enough for each action to read, maintaining focused eye contact with her reflection. The camera passes behind a stylist’s shoulder and returns to the mirror as the final makeup light clicks on. The circular vanity bulb blooms into a full white flare. The white flare becomes a professional studio strobe. SHOT 4 — STUDIO PERFORMANCE AND THE RUSH OUT Reveal @image5 at the instant of the studio flash. The model performs two strong connected movements in the same crimson couture dress: a sharp turn that sends the fabric panel outward, followed by a controlled stop facing the camera. Use a fast low orbit synchronized with the fabric. The industrial fan drives the dress and hair motion. A photographer’s camera and black flags cross the foreground, creating strong depth and visual rhythm. At the next strobe flash, a black travel coat is thrown around her shoulders off-screen. She is already walking rapidly away from the set as the flash clears. The camera retreats down the studio corridor in front of her. She checks the time without revealing readable screen text and increases her pace. An assistant opens a black car door. She enters without stopping. The car door closes directly against the lens, cutting the image to black. SHOT 5 — CAR-DOOR MATCH CUT INTO THE RUNWAY The black frame opens as a backstage runway curtain parts in the same direction as the car door. The model emerges in the black sculptural runway gown from @image6 . The wardrobe change has occurred entirely backstage and off-screen. Use a low frontal camera traveling backward along the runway as she advances with powerful, precise steps. Halfway down the runway, arc quickly from a low frontal view into a close side profile, capturing the gown’s weight, silver shoulder detail and controlled arm swing. Audience cameras flash in irregular bursts. Her expression remains composed and professional. At the strongest flash, the runway disappears into white light. SHOT 6 — OUTDOOR CAMPAIGN AND RIDE HOME The white runway flash becomes a direct golden sun flare on the outdoor rooftop. Reveal @image7 with the model now wearing the ivory trench and silver-gray tailored outfit. The change happens between locations, never as a visible morph. A strong coastal gust catches the trench as she pivots toward the camera. Use a rapid low push-in followed by a controlled half-orbit. Her hair, trench panels and nearby diffusion fabric move from the same wind source. She takes two purposeful steps and lands in one decisive advertising pose as the photographer fires. Do not hold the pose. An off-screen assistant signals completion and she immediately turns toward a waiting black car. The ivory trench sweeps across the lens. When it clears, she is seated inside the same car at night, still wearing the ivory trench. City lights travel rapidly across her face and the window. She leans her head briefly against the glass. Her car-window reflection becomes her bathroom-mirror reflection. SHOT 7 — REMOVE THE DAY AND COMPLETE THE CYCLE In the bathroom mirror, she wipes the final trace of makeup from one eye. Use one rapid sequence of physically connected actions: cleansing water over her face, towel crossing the lens, hair released and becoming slightly damp. The towel lowers to reveal her back in the charcoal-gray pajamas from the morning. She walks directly into the dark bedroom and falls backward onto the white bed with genuine exhaustion. Match her landing position to @image9 . The camera follows down to mattress height. She reaches toward the bedside lamp, switches it off and closes her eyes. End in the same composition and physical position as the opening, now under deep blue night light. Hold darkness only briefly. NATURAL PACING Let the AI determine the natural internal duration of each shot within the total 15 seconds. Keep the morning routine and final bedtime concise. Give the strongest visual emphasis to the street walk, studio fabric movement, runway walk and wind-driven outdoor campaign. Maintain continuous forward momentum. Use short readable actions rather than random rapid cuts. CAMERA LANGUAGE Use: rapid facial push-in -> bedding wipe -> mirror perspective -> water-on-lens transition -> low street tracking -> backward walking shot -> vehicle wipe -> vanity orbit -> light flare -> low studio orbit -> corridor retreat -> car-door cut -> runway tracking -> profile arc -> flash transition -> rooftop push-in and half-orbit -> coat wipe -> intimate car tracking -> reflection match cut -> mattress-level final descent. Every movement must reveal action, space or emotional state. Do not use camera motion merely as decoration. PHYSICAL REALISM Preserve realistic anatomy, foot contact, walking cadence, fabric weight, hair inertia, water droplets, mirror reflections, car-window reflections, door movement and lighting direction. Background people remain secondary and must never duplicate the main model. AUDIO Use a fast alarm vibration, running tap water, towel movement, street traffic, boot steps, vanity tools, hair spray, camera shutters, studio fan, hurried corridor footsteps, car-door impact, runway heels, audience camera bursts, rooftop wind, muted car ambience and final running water. No dialogue, narration or lyrical music. A restrained electronic fashion beat may build from the city walk through the outdoor shoot, then stop cleanly when the bedside lamp switches off. The finished film must feel like one relentless but coherent professional day, not a collection of disconnected fashion images, slow beauty shots or generic static image animation.

**Main Subject:** Young Korean woman , 24, naturally attractive with an understated everyday appearance. Long dark hair loosely tied back, wispy strands around her face, realistic skin texture, minimal makeup. Wearing an oversized faded peach T-shirt, loose charcoal lounge pants, white socks, and simple house slippers. Thin silver necklace. Preserve her exact identity, facial features, hairstyle, and appearance throughout. **Location:** Modest older Seoul apartment on a quiet weekend morning. Narrow entryway, worn wooden floors, small shoe rack, cluttered kitchen, low dining table, faded curtains, old refrigerator, cardboard boxes and an open window overlooking dense apartment buildings. Soft natural morning light. **Style:** Ultra-realistic early-2000s Sony MiniDV home video. Casual family recording, completely unstaged and imperfect. Handheld shake, awkward framing, autofocus hunting, exposure shifts near windows, faded colors, soft contrast, DV compression, slight motion blur and microphone noise. No stabilization or modern cinematic movement. **00:00–00:03:** She hears the apartment buzzer while sitting on the floor sorting through some clothes. She looks toward the door, confused, then gets up and walks into the hallway. **00:03–00:06:** She opens the door and receives a medium-sized cardboard package from a delivery person. She looks at the box, then briefly toward the camera with a curious expression. **00:06–00:09:** She carries it inside and places it on the floor. Sitting beside it, she struggles slightly with the packing tape using her fingernail before finally tearing it open. **00:09–00:12:** She looks inside and suddenly smiles with genuine surprise. She pulls out the unexpected item and turns it toward the camera, laughing while someone behind the camera reacts. **00:12–00:15:** She examines it happily for a moment, then realizes the camera is still recording. She gives the operator an amused look and says something off-camera before the recording abruptly ends. **Audio:** Natural apartment ambience only—door buzzer, hallway footsteps, cardboard scraping, packing tape tearing, distant traffic, refrigerator hum, faint neighbors and casual off-camera conversation. No music or narration. **Goal:** Feel like a forgotten personal MiniDV clip from the early 2000s—an insignificant delivery that accidentally captured a genuine little moment of surprise. Warm, mundane, spontaneous, intimate and imperfect.

15-second, 16:9 horizontal video. A fusion of live-action footage inside a dark cozy bedroom and hand-drawn glowing animation. A lived-in bedroom illuminated only by a warm bedside lamp, faint fairy lights, and the glow of a laptop screen. Clothes, books, cables, cups, and everyday objects remain naturally scattered around the room. The footage feels like spontaneous one-handed smartphone recording, with subtle hand shake, low-light sensor noise, occasional focus hunting, motion blur, rolling exposure shifts, imperfect framing, and natural auto-exposure fluctuations. 0–3 seconds: A finger touches the surface of a glass of water on the desk. A thin glowing turquoise hand-drawn line leaks out from beneath the fingertip and crawls across the desk like liquid ink. The line curls upward, breaks apart into rough scribbled fragments, then continuously reshapes into a tiny glowing jellyfish floating above the desk. The jellyfish suddenly drifts toward a nearby bookshelf. The animation moves first, then the handheld camera reacts a fraction of a second later, quickly panning and stepping after it. 3–6 seconds: The jellyfish slips between two books. Its tentacles stretch into loose scribbles, its body collapses into uneven colored strokes, and the entire drawing continuously morphs into a small orange fox made from glowing crayon lines. The fox jumps down from the shelf, runs across the floor, then dives beneath the bed. The animation escapes first while the camera hesitates, tilts downward, then follows slightly late with imperfect handheld movement. 6–10 seconds: Under the bed, the orange fox stretches unnaturally long. Its legs dissolve into restless pencil strokes while its body twists and continuously reshapes into a green lizard. As it crawls out from the opposite side, the lizard's body fragments into fluttering chalk marks and morphs into a flock of tiny purple birds. The birds spiral around the room before merging together, their rough strokes tangling and continuously transforming into a large floating dragon made entirely from hand-drawn lines. The dragon rushes toward the bedroom door first, forcing the camera operator to stand, turn, and chase after it with delayed, shaky smartphone movement. 10–13 seconds: The dragon crashes into a hanging jacket beside the door without physically damaging it. Its body scatters into hundreds of glowing sketch marks that crawl across the fabric. The jacket suddenly develops two rough cartoon arms and a smiling face. Its smile stretches wider as the colored strokes break apart, twist, overlap, and continuously morph the friendly jacket character into a strange fuzzy monster with oversized eyes and crooked teeth. The monster jumps toward the ceiling first, while the camera jerks upward a moment later and struggles briefly to regain focus. 13–15 seconds: The monster splashes across the ceiling as loose crayon strokes. The strokes spread rapidly across the surface, curl together, and continuously reshape into a gigantic hand-drawn moon face covering most of the ceiling. Its eyes open, blink once, then its mouth stretches into a playful grin. The face suddenly looks toward the bedroom door as the last glowing strokes ripple across the walls. The animation reacts first, then the camera slowly tilts and drifts after its gaze with a noticeably delayed handheld response. All animated elements must feel physically integrated into the live-action room while remaining unmistakably handmade. Use rough crayon, chalk, colored-pencil, dry-brush, and scribbled textures, irregular outlines, visible paper-like grain within the strokes, imperfect fills, fuzzy edges, uneven opacity, and subtle frame-by-frame line wobble. Transformations must happen visibly through collapsing, stretching, fragmenting, smearing, tangling, and reshaping strokes, never by making one object disappear and replacing it with another. Maintain one continuous take inside the same room with no cuts or scene transitions. The animated creatures always initiate movement first, while the person filming reacts slightly late and follows them naturally with one-handed smartphone movement. Preserve handheld shake, low-light noise, focus breathing, imperfect autofocus, motion blur, exposure fluctuations, accidental framing, and human reaction time. Environmental audio only: quiet room tone, footsteps, fabric movement, floor creaks, distant outdoor ambience, refrigerator hum, subtle breathing, and natural handling noise from the phone. No overly precise 3D CGI, excessively smooth vector lines, neon-tube-like visuals, perfectly stabilized gimbal footage, overly cinematic lighting, polished advertising-style compositions, subtitles, logos, or background music.

A young Japanese woman wearing an elegant feminine summer dress with a soft floral pattern and flowing fabric. She walks casually through a charming Japanese city street on a pleasant afternoon, holding a cold drink. 0–5 sec: She walks naturally while looking around, her hair and dress gently moving in the breeze. Her female friend films her handheld from the side. 5–10 sec: She stops near a small café, takes a sip of her drink, checks her phone, then looks around with a soft, cute smile. 10–15 sec: She notices the camera, gives a sweet little smile, slightly raises her drink and says naturally in Japanese: 「今日は自分の時間。」 Then she turns back and continues walking. Maintain the same face, hairstyle, outfit, body proportions, and identity throughout the video.

A serene high-angle scene of a young woman with long dark hair lying peacefully on a lush, moss-covered riverbank beside a flowing stream. She wears a dark camouflage knitted sweater. Her hand is initially raised toward leafy tree branches hanging overhead, slowly descending down to her side as she gazes softly upward with a tranquil expression. Sunlight filters through the leafy canopy, casting shimmering highlights across the rippling dark water. The artwork features a detailed, cinematic anime digital art style with soft atmospheric lighting, rich earthy green and brown tones, and smooth, gentle movement.

**Prompt:** *A cinematic magical-realism sequence set on a quiet European-style city street during a gentle rainfall at blue hour. A beautiful young woman in a light pastel raincoat walks calmly through the rain. Curious, she raises her hand and catches a single raindrop. The instant it touches her fingertips, it transforms into a delicate blooming flower. As she continues walking, every raindrop that lands on her or touches the ground blossoms into vibrant wildflowers, roses, daisies, cherry blossoms, and colourful petals. Within seconds, the once-grey street transforms into a breathtaking floral paradise. Flowers bloom naturally from the wet pavement, climb walls, decorate lamp posts, and spread across rooftops. Soft petals swirl gracefully in the wind while warm golden streetlights reflect on the rain-soaked road, creating a dreamy, enchanting atmosphere.* *The camera opens with an extreme close-up of the raindrop touching her fingertips, then smoothly transitions into a slow cinematic tracking shot following her as flowers bloom behind every step. Include realistic character animation, subtle facial expressions of wonder, natural walking motion, flowing hair and clothing physics, volumetric rainfall, realistic water splashes, soft fog, depth of field, cinematic lens flares, HDR lighting, global illumination, highly detailed flower growth animation, seamless motion, ultra-realistic textures, magical realism, premium film quality, 4K, 24fps, masterpiece, emotionally uplifting, visually breathtaking.* **Negative Prompt:** *Low quality, blurry, low resolution, cartoon, anime, CGI look, plastic skin, unrealistic flowers, duplicated flowers, flickering, jitter, unstable motion, deformed anatomy, extra fingers, bad hands, distorted face, oversaturated colours, flat lighting, frame glitches, artifacts, text, subtitles, watermark, logo, noise, compression, camera shake, abrupt transitions, unrealistic physics.*

Create a 15-second original hand-drawn anime-style image-to-video sequence using no more than 5 coherent shots. REFERENCE RULES Use Keyframe A as the exact opening-frame reference. Use Keyframe B as the character-action, dog-running and courtyard-geography reference. Use Keyframe C as the required final-frame reference. Use the separate character references to preserve the exact boy, grandfather, grandmother and dog throughout the video. Preserve the hand-painted watercolor and gouache backgrounds, delicate pencil linework, traditional cel shading, warm rural summer palette and original character designs. Do not turn the supplied illustrations into photorealism, 3D animation or glossy modern CGI. ENTITY CONTINUITY Maintain exactly: - one 16-year-old boy; - one grandfather; - one grandmother; - one golden-tan Shiba-mix dog with one red collar; - one blue morning-glory uchiwa fan. The grandparents remain safely near the engawa throughout the chase. The fan follows one continuous state chain: one fan resting on the table -> the same fan held by the dog’s mouth -> the same fan released into the air -> the boy catches the same fan -> the grandfather holds the same fan in the final shot. Never show two fans at the same time. CORE COMEDIC MECHANISM The family prepares to eat watermelon -> the dog notices the fan -> the dog suddenly steals it -> the boy launches into a fast courtyard chase -> the dog releases the fan upward -> the boy catches it -> the grandfather calmly takes the recovered fan and uses it to cool everyone. The humor comes from the contrast between the boy’s energetic pursuit, the dog’s playful confidence and the grandparents’ relaxed amusement. SHOT 1 — WATERMELON HOOK AND THE DOG’S DECISION Begin exactly from Keyframe A. Start with a rapid macro glide over condensation rolling down a vivid red watermelon slice. The grandmother’s hand places the slice onto a plate. The camera continues past the fruit toward the boy, who leans forward eagerly. Before he can take the slice, rack focus behind his arm to the dog’s eye watching the blue morning-glory fan. The dog’s ears rise. It looks at the boy, looks at the fan, then makes one quick decision. Use a fast controlled push-in on the dog’s face, followed immediately by the dog springing forward and grabbing the bamboo edge of the fan. The boy’s hand reaches the empty space a fraction too late. The dog bolts out of frame, pulling the camera into a sharp motivated whip-pan. Start the comic action almost immediately. Do not linger on the food setup. SHOT 2 — LOW DOG-EYE CHASE Continue the whip-pan into a very low tracking shot beside the running dog. The dog races across the courtyard with the same fan held securely but gently in its mouth. Its paws kick up small dry leaves. Its ears, curled tail and red collar react naturally to the running motion. The boy jumps down from the engawa and chases after it. His open white overshirt, hair and shorts respond with believable weight and overlapping hand-drawn motion. The camera accelerates alongside the dog, passing close to a metal wash basin, stepping stones and potted flowers. These foreground objects create strong multiplane parallax. The dog curves around the persimmon tree. The camera performs a fast controlled half-orbit around the trunk and reveals the boy cutting across the opposite side to intercept it. The grandparents remain visible beneath the veranda in the deep background, reacting with restrained laughter. SHOT 3 — TABLE DODGE AND REVERSAL Use the boy’s white overshirt sweeping across the lens as a natural transition. The dog ducks beneath the low table and emerges from the other side. The boy stops himself just before colliding with the table, plants one hand on its edge and swings around it. Nothing on the table spills. The watermelon, peaches and cups shake slightly from the sudden movement but remain in place. The camera travels backward close to the ground as the dog runs directly toward it. The boy gains on the dog and reaches for the fan. At the final instant, the dog changes direction around the grandmother’s house slippers, playfully avoiding his hand. The boy pivots and continues rather than falling. Keep the action readable, agile and joyful, not chaotic slapstick. SHOT 4 — THE FLYING FAN PAYOFF Track sideways as the dog crosses a sunlit patch beneath the persimmon tree. The boy finally reaches close enough that the dog playfully releases the fan. The same fan spins once upward through a shaft of sunlight. Briefly follow the rotating fan from below. Sunlight flashes across the blue morning-glory painting while green leaves sweep behind it. Use a very short speed ramp only at the top of the fan’s arc. The boy steps beneath it and catches the bamboo handle cleanly with one hand. The dog immediately turns back, performs a playful bow and barks once, clearly satisfied with the game. The boy looks at the fan, then at the dog. His mock frustration breaks into laughter. Follow the fan downward as the boy lowers it, using its round surface to wipe naturally into the final scene. SHOT 5 — RETURN TO THE SHADE As the fan clears the lens, reveal the boy dropping onto the engawa between his grandparents. The grandfather calmly takes the same recovered fan from the boy and begins fanning the group with broad, rhythmic movements. The grandmother hands the boy the watermelon slice he missed at the beginning. He takes one large refreshing bite. The dog circles once at his feet, lies belly-down and rests its head on his ankle. A stronger summer breeze arrives at the same moment: the wind chime swings, the bamboo blind lifts, laundry billows, persimmon leaves turn in the sunlight and everyone’s clothing responds naturally. The camera moves backward along the veranda and rises into the composition of Keyframe C. End with the entire family beneath the shaded eaves, the grandfather fanning them, the boy eating watermelon, the grandmother smiling and the exhausted dog resting at his feet. Hold the final image only briefly. The ending is warm and satisfying without losing the energetic rhythm. NATURAL PACING Let the AI determine the exact duration of each shot within the total 15 seconds. Keep the fruit setup extremely short, give most of the runtime to the fan theft and physical chase, and reserve only the final few seconds for the warm family payoff. Use fast subject-driven camera movement rather than excessive cutting. CAMERA LANGUAGE Use: macro food glide -> rapid rack focus -> push-in -> whip-pan -> low dog-eye tracking -> tree-trunk half-orbit -> foreground clothing wipe -> backward ground tracking -> upward fan follow -> short speed ramp -> fan-surface wipe -> veranda pullback and crane rise. All camera moves must be motivated by the dog, boy, fan or physical foreground objects. Preserve readable courtyard geography and continuous screen direction. ANIMATION DIRECTION Fluid theatrical 2D character animation with strong key poses, clean silhouettes, expressive timing, realistic weight and controlled squash-and-stretch only on the dog’s running motion. Use hand-drawn overlapping motion in the boy’s hair, overshirt, dog’s ears, curled tail, laundry, bamboo blind, leaves and spinning fan. Keep faces, hands, paws, clothing shapes, linework and background painting stable between frames. AUDIO Dense midsummer cicadas, a faint wind chime, ice shifting in the barley-tea pitcher, dog paws against packed earth, dry leaves, fabric movement and the bamboo fan cutting through the air. Add one short dog bark, the boy’s breathless laughter and the grandparents’ restrained chuckling. No dialogue, narration, dramatic score or exaggerated cartoon sound effects. The finished video must feel like a lively miniature summer story, not slow scenic parallax, a generic family montage, static characters blinking, or disconnected anime shots.

A young traveler hikes through a lush mountain forest, crosses a wooden bridge over a flowing stream, and reaches a stunning waterfall hidden among moss-covered rocks. Ultra-realistic travel documentary, authentic hiking movement, realistic environmental physics, natural English lip-sync, bright daytime lighting, immersive forest ambience, and seamless story continuity throughout.

15-second handheld phone video, one shot, no cuts, 16:9. The footage IS the phone's view the phone itself is never visible, no phone or screen in shot. SETTING: A freezing, pitch-black Christmas night in 1776 on a wide, fast-moving river clogged with broken slabs of grey river ice. Heavy overcast sky, no moon, the only light coming from a single swinging iron lantern hung at the front of the boat and a second dim lantern further back, both throwing shaky warm-orange pools across the water and faces while everything beyond falls into deep blue-black darkness. Thin sleet drifts through the lantern light. The far riverbank is barely visible as a black mass with a faint dusting of snow. THE BOAT: A large flat-bottomed wooden Durham cargo boat sitting low and heavy in the water, its plank hull dark and wet, crammed shoulder to shoulder with ten to twelve Continental Army soldiers in worn dark-blue and brown wool coats, tricorn hats and a few knit caps, some with rags wrapped around their hands and feet, muskets and a small cannon wheel visible among them. Their faces are pale, wet, and exhausted, breath steaming heavily in the cold. Two soldiers pole and push slabs of ice away from the bow with long oars; two more row hard at the sides. THE OFFICER: Near the bow stands a tall, imposing officer in a long dark cloak and a black tricorn hat, one boot up on the gunwale, holding steady against the rocking, looking forward toward the far shore — calm and rigid while everyone around him strains and shivers. CAMERA & MOVEMENT: Filmed from low inside the boat, roughly at bench height among the soldiers, as if held by a soldier crouched in the crush. Violent handheld shake with every jolt of the hull against ice, constant micro-jitter, the operator's own breath fogging across the lens in bursts. Autofocus hunts and pulses between the swinging lantern flame, the standing officer's back, and the black churning water sliding past. Oars and elbows swing past the lens. A wet sheen and stray sleet-drops speckle the lens. The view lurches as a large ice slab grinds down the length of the hull and the boat pitches; a soldier braces a hand against the camera-holder's shoulder, and the operator swings the view up briefly toward the officer at the bow, then back down to the ice-choked water. AUDIO: on-phone mic, overdriven and slightly distorting in the cold wind. Thick slabs of ice scraping and cracking hard against the wooden hull, oars sloshing and knocking in the locks, low urgent hushed male voices giving quiet orders, teeth chattering, heavy breathing, wind buffeting and clipping the mic, the fast river hissing and gurgling past the boat. STYLE: Photoreal, physically accurate historical footage, authentic raw handheld phone look in near-darkness, natural lantern lighting with deep shadows and heavy film-grain noise in the blacks, realistic water and ice physics, real human weight and shivering, consistent low-light exposure. No CGI look, no slow motion, no music, no text, no timestamp, no logos, no modern objects, no phone visible in frame

Main Subject: Young Korean woman, 23, natural everyday appearance, shoulder-length dark brown hair worn loose with soft windswept bangs, realistic skin texture, minimal makeup, calm and thoughtful personality. Wearing a light beige trench coat over a white knit top, straight-fit blue jeans, white sneakers, a small brown leather shoulder bag, and a silver wristwatch. Maintain consistent identity, clothing, hairstyle, and appearance throughout. Location: A small coastal train station in Busan during a bright autumn morning. Quiet outdoor platform overlooking the sea, weathered benches, yellow safety lines, overhead power cables, station signs in Korean, distant green hills, and gentle ocean waves beyond the tracks. Only a few commuters are waiting. Visual Style: Ultra-realistic documentary realism. Authentic slice-of-life, genuine candid behaviour, natural body language, soft autumn sunlight, rich environmental details. Camera Style: Early-2000s Sony MiniDV camcorder. Heavy handheld shake, imperfect framing, autofocus hunting between passing trains and her face, exposure pumping under changing sunlight, faded colours, soft contrast, slight motion blur, MiniDV compression artifacts, no stabilization. 00:00–00:03 She stands near the platform edge, holding a warm takeaway coffee while watching the distant railway tracks. The camera briefly focuses on the station sign before finding her face. 00:03–00:06 A cool sea breeze blows through her hair as she checks the train schedule and quietly smiles. The operator accidentally zooms in too far before correcting. 00:06–00:09 She sits on a wooden bench, taking a sip of coffee while looking toward the ocean. Autofocus shifts between her profile and the sparkling water behind her. 00:09–00:12 The train slowly arrives with a soft metallic rumble. She stands, adjusts her shoulder bag, and steps toward the open doors. 00:12–00:15 Before boarding, she notices the camcorder, gives a small genuine wave and smile, then enters the train. The doors begin to close as the camera drifts downward, and the recording abruptly cuts to black. Audio: Natural ambience only–ocean waves, seagulls, light wind, distant station announcements, footsteps, train brakes, doors opening, quiet commuter conversations. No music. No narration. Goal: A forgotten MiniDV home video from autumn 2004, capturing a peaceful morning at a small coastal train station in Busan–simple, candid, nostalgic, and completely believable.

The shot opens tight on a beautiful woman face, wind tearing at her coat as she stands on a jagged cliff. The camera eases over her shoulder and tilts down, following her line of sight. We descend past the crumbling lip into the void, the frame widening to reveal cliffs dropping away. The ocean roars into view, violent surf detonating against rock. The move continues into a high overhead, locking into an aerial top-down that frames the lone figure at the precipice as waves explode below.

Speeder chase across a cliff city (single continuous shot) From a monumental cliffside city carved into stone, the camera dives toward a tiny streak of light ripping along a narrow ledge-road. Lock-on: a speeder hugging the wall at insane speed. The camera slingshots ahead, whips back, then drops tight to the rear thrusters: heat haze, grit snapping off the ledge, warning lights flashing. A collapsing balcony rains debris; the rider snaps a last-inch swerve under a falling arch, then threads through hanging laundry lines and open windows in one fluid line. The camera darts through the same openings, staying glued to the motion. One final bend and sudden calm: the camera blasts outward into a reveal of the city opening onto a boundless waterfall-fed valley, mist turning into rainbow.

REFERENCE IMAGE RULES Use the supplied character portrait as the absolute identity and wardrobe reference. Use the supplied sleeping-in-the-forest image as the opening-frame reference. Use the forest environment image to preserve the geography, vegetation, dawn mist and cool blue-green lighting during the first part of the sequence. Use the river environment and final river image as the destination and last-frame reference. The video must begin from the supplied forest opening image and arrive naturally at the supplied river composition. Do not simply dissolve or morph between the two images. The woman must physically travel through a believable connected forest space. TONE AND ENERGY Cinematic magical realism with urgent physical movement, mounting tension and a warm emotional reversal. The opening must hook the viewer immediately. Start the main action within the first second. Use a fast, escalating rhythm: sudden awakening -> mysterious animal trigger -> rapid forward movement -> physical stumble and downhill slide -> explosive river reveal -> brief moment of wonder and relief. The forest initially feels as if it is closing in around her, but the final reveal shows that it has been guiding her. This is tense and adventurous, not horror, not slow meditation and not a dreamy fashion film. CHARACTER CONTINUITY Maintain the exact same young woman from all supplied reference images: identical face, age, body, shoulder-length dark brown hair, cream knitted cardigan, muted sage-green dress, brown lace-up ankle boots and olive canvas crossbody bag. The bag strap remains attached from her left shoulder to her right hip throughout the entire sequence. Her clothing progressively becomes slightly damp and muddy, but never changes design or color. Her emotional progression is: startled confusion -> frightened urgency -> physical determination -> sudden wonder -> visible relief. ENTITY CONTINUITY Maintain exactly one woman and one white rabbit during the forest journey. The same white rabbit reappears at the river. Exactly one young fawn and two small woodland birds may appear only in the final river section. No animal duplicates, replacements or spontaneous materialization. Every animal must enter the frame through a physically visible route. CORE CAUSE-AND-EFFECT CHAIN A branch snaps and wakes the woman -> the same white rabbit darts across her vision -> she follows it through the forest -> she stumbles and slides down a damp slope -> the rabbit passes through a curtain of ferns -> she pushes through the same ferns -> the river is revealed -> the rabbit calmly drinks beside the water -> a fawn and two birds appear -> she understands that the rabbit was guiding her. SHOT 1 — VIOLENT AWAKENING AND IMMEDIATE HOOK Begin exactly from the supplied opening image of the woman lying on wet moss. Extreme macro focus on a dew droplet falling from a fern and striking her cheek. A branch suddenly snaps off-screen. Her eyes open sharply and she inhales. She rolls onto one elbow and rises quickly, still disoriented. At the same instant, the camera rapidly pulls backward from the close-up into a low, slightly tilted medium shot, revealing dense trees and foreground ferns surrounding her. A white rabbit suddenly darts through the near foreground, crossing very close to the lens. The movement creates a natural whip-pan. The camera catches the rabbit landing briefly on a moss-covered root. It looks back once. The woman locks eyes with it. Before she can react, the rabbit runs deeper into the forest. Do not linger on the awakening. The rabbit must trigger forward movement almost immediately. SHOT 2 — FAST FOREST PURSUIT Use the rabbit’s whip-pan as the transition into an energetic low tracking shot beside the woman’s boots. She pushes herself upright and runs after the rabbit through the narrow opening between the trees. The camera races alongside her at knee height, extremely close to wet ferns and exposed roots. Foreground vegetation creates strong parallax and repeatedly sweeps past the lens. Cut within the same movement to an FPV-style camera traveling backward in front of her as she runs toward it. Her cardigan, dress hem, hair and bag react naturally to her speed. Her boots strike wet leaves and throw small droplets and pieces of moss backward. The rabbit remains visible only in brief readable glimpses ahead of her: crossing behind one tree, appearing on a root, then moving forward again. It never teleports. The forest feels compressed and urgent. Tree trunks pass close on both sides while the camera maintains her forward screen direction. SHOT 3 — STUMBLE, DROP AND SPEED RAMP Transition behind a dark tree trunk crossing the entire frame. The camera emerges into a fast side-tracking shot as her boot catches beneath a thick root. She stumbles forward, grabs a flexible branch and briefly swings around it rather than falling flat. The branch bends realistically under her weight and shakes loose a shower of dew. She regains one step, but the damp ground collapses beneath her next foot. The camera drops with her as she slides several meters down a fern-covered slope. Use one very brief controlled speed ramp at the instant her foot loses contact, then return immediately to real-time motion. She protects her face with one arm, turns onto her side and stops against soft moss. Her bag remains attached to her body. Before she can fully recover, the white rabbit rushes past the lower edge of the slope and disappears through a bright wall of ferns. The woman hears the river for the first time. She looks up sharply toward the sound, gets to her feet and charges after it. Keep the stumble physically believable and exciting, not comedic and not prolonged. SHOT 4 — FERN WIPE AND EXPLOSIVE RIVER REVEAL Track tightly behind her shoulder as she runs toward the bright ferns. She raises both arms to protect her face and pushes forcefully through them. The ferns sweep completely across the lens, creating a motivated full-frame transition. As the foliage clears, the camera bursts through with her and performs a fast controlled 180-degree arc: begin facing her exhausted expression while moving backward, then swing behind her shoulder and rise to reveal the crystal-clear river filling the space ahead. The enclosed cool forest suddenly opens into warm golden morning light, moving water and a wide horizon between the trees. She tries to stop on the wet bank, slides one short step, drops onto one knee and catches herself with one hand against the river stones. Her hand touches the water, producing one clear splash and realistic concentric ripples. The camera’s aggressive movement settles rapidly into a smooth close orbit as the sound and space open around her. SHOT 5 — ANIMAL REVEAL AND EMOTIONAL REVERSAL Continue from the orbit without a hard cut. The same white rabbit steps visibly from the riverside grass and calmly drinks near her hand. She initially tenses, recognizing it. Rack focus across the moving water to exactly one young fawn emerging from behind the pale tree on the opposite bank. The fawn takes two natural steps, stops and meets her gaze. Exactly two small birds descend into frame and land on a branch above the fawn, causing the branch to move slightly under their weight. The animals remain realistic and calm. Their spirit-like quality comes only from intelligent eye lines, coordinated stillness and soft natural rim light. They do not glow. The woman looks from the rabbit to the fawn and understands that she was being led rather than chased. Push quickly into an intimate close-up as her rapid breathing slows, her clenched jaw releases and her shoulders drop. A small genuine smile begins, restrained and emotionally believable. Complete the movement with a brisk crane rise and pullback toward the supplied final-frame composition: the woman beside the clear river, the same rabbit near her, the fawn across the water, two birds above and the sunlit path visible behind them. Hold the final composition only briefly, long enough for the emotional reversal to register without slowing the overall rhythm. CAMERA AND EDITING FINISH Use strong spatial travel rather than static zooming. Camera progression: rapid macro pullback -> whip-pan -> low lateral tracking -> backward FPV tracking -> trunk wipe -> falling camera with speed ramp -> foliage wipe -> controlled 180-degree reveal -> close orbit -> rack focus -> crane pullback. Every transition must be motivated by a physical foreground element or continuous character movement. Maintain the woman’s forward screen direction and continuous downhill geography. Use realistic handheld energy during the forest pursuit, then transition rapidly into controlled cinematic stability at the river. Allow the AI to determine the natural duration of each shot within the total 15 seconds. Keep the awakening extremely short, give the forest pursuit and fall most of the runtime, and reserve only the final few seconds for the river and emotional payoff. AUDIO No dialogue and no narration. Use a sharp branch snap, sudden inhale, rapid footsteps, wet leaves, fabric movement, breaking twigs, sliding soil and increasingly loud breathing. Introduce the sound of running water immediately after the downhill slide. At the river reveal, let the enclosed forest sound open suddenly into flowing water, wingbeats, light birdsong and a warm breeze. No slow sentimental music, horror stingers or magical chimes. The result must feel like an energetic miniature adventure with a clear reversal, not generic still-image animation, slow parallax, a woman calmly walking through a forest or a montage of unrelated scenic shots.

Main Subject: Young Korean woman 24, naturally attractive, realistic skin, minimal makeup, long dark hair loosely tied back. Wearing a loose cream cardigan over a faded red T-shirt, relaxed blue jeans, white sneakers, and a thin silver necklace. Preserve her exact identity, facial features, hairstyle, and appearance throughout. Location: Small old-fashioned Seoul neighborhood arcade in the early evening. Dim fluorescent lighting, rows of aging arcade cabinets, colorful game screens, scratched floors, plastic stools, faded posters, coin machines, tangled cables, and a few local teenagers playing in the background. Warm street light occasionally spills through the entrance. Style: Ultra-realistic early-2000s Sony MiniDV home video. Authentic, candid and imperfect. Heavy handheld shake, awkward framing, autofocus hunting between bright game screens and faces, exposure pumping, faded colors, soft contrast, DV compression, slight motion blur and microphone distortion. No stabilization or modern cinematic movement. 00:00–00:03: She pushes open the arcade door and steps inside. The camera follows her slightly late, briefly overexposing from the bright street behind her before adjusting to the dark interior. 00:03–00:06: She walks past several machines, looking at the screens and deciding what to play. The camera wanders toward different games before finding her again. 00:06–00:09: She sits at an old fighting-game cabinet and starts playing. She becomes surprisingly competitive, leaning toward the screen and rapidly pressing the buttons while the camera shakes behind her. 00:09–00:12: She wins a round and immediately celebrates with a small fist pump. She turns toward the camera with a proud grin, then loses the next round almost instantly and bursts out laughing. 00:12–00:15: She stands up and walks toward another machine. As she passes, the camera accidentally zooms tightly onto a glowing arcade screen, autofocus struggles against the pixels, and the recording abruptly cuts to black. Audio: Authentic arcade ambience only—button clicks, joystick sounds, electronic game noises, coin mechanisms, fluorescent buzzing, distant conversations, footsteps and occasional laughter. No added music, narration, or cinematic sound effects. Goal: Feel like a forgotten personal MiniDV recording from an ordinary evening at a small Korean arcade in the early 2000s. The humor should come naturally from her reactions and the imperfect camera operation. Nostalgic, intimate, spontaneous and completely unpolished.

modern street culture shot like a premium sneaker film: a concrete skatepark at golden hour, a rooftop streetball court at dusk, a graffiti-covered subway underpass at night; warm orange as the through-color; one young skater in a bright orange hoodie and backwards black cap, his skateboard with him in every scene; three scene changes, and each one is a single continuous action that changes its body mid-motion, never a cut. Action: he fires the final arc of orange spray onto a huge abstract mural and tosses the can high into the air; the camera rises with the tumbling can, and at the top of its arc, without the flight ever pausing, the can becomes an orange basketball that falls into the dusk of a rooftop court where he catches it, board under his arm; he dribbles once and shoots, the ball spinning off the rim toward the lens; the camera closes in until the spinning pebbled surface dominates the frame, and without the spin ever stopping it becomes a spinning vinyl record; the camera pulls back to reveal the graffiti-covered underpass, his hand on the record; he scratches, the soundtrack bending under his fingers, and the orange glare on the groove stretches into a streak of light; without the stretching ever stopping, the streak becomes the lengthening shower of orange sparks from his skateboard grinding a metal rail; the camera pulls back along the spark trail into the night skatepark, he pops off the rail, rolls out and stops in a held pose under the sodium floodlights. Camera: a low tracking shot beside him at the mural; a rise following the tossed can that turns into a fall following the ball; a slow arc as he shoots; a push into the spinning ball that becomes a pull-back from the spinning vinyl; a locked shot on his scratching hand; a push along the stretching light streak that becomes a pull-back along the spark trail; a tracking settle as he rolls to a stop; no hard cuts anywhere, every scene change happens at the closest point of one continuous camera breath. Timing: 0-3s: golden hour, the final arc of spray on the mural, he tosses the can up; 3-7s: the can tumbles through the air and at the apex of its arc becomes the falling basketball, caught on the rooftop court; 7-9s: one dribble and the shot, the ball spinning off the rim toward the lens; 9-12s: the spinning ball fills the frame and, never stopping its spin, becomes the spinning vinyl, the camera pulling back into the underpass; 12-14s: his hand scratches the record, the music bends, the orange glare on the groove begins to stretch; 14-17s: the stretching streak of light becomes the lengthening shower of sparks from his board grinding the rail, the camera pulling back along it; 17-20s: night skatepark, he pops off the rail, rolls out with an ease-out and holds a still pose for the final two seconds. Audio: (a boom-bap hip-hop instrumental running through the whole film, a hard snare hit exactly on each body swap; when he scratches, the soundtrack itself bends, revealing it was his record all along); <spray can hiss>, <the rattle of the tumbling can>, <a slap as he catches the ball>, <the ring of the rim>, <vinyl scratch>, <crackling grind sparks>, <a clean snap as he lands off the rail>, then the beat falls away to quiet night city hum for the held pose; no dialogue. Constraints: the flight never pauses while the can becomes the ball; the spin never stops while the ball becomes the record; the stretching never stops while the light streak becomes the sparks; no cuts, no wipes, no occlusion transitions, no whip pans, no fades, no cross dissolves, no random morphing; the same single young man in every scene with identical orange hoodie, backwards black cap and skateboard, exactly two arms and two hands, no other people anywhere; all graffiti is abstract shapes with no readable letters or words; no brand logos; no on-screen text, no subtitles.

A 15-second photorealistic smartphone selfie vlog that looks exactly like a real mobile phone recording. A young woman (image = her face and hair) is exploring the quiet streets of Kyoto, Japan on a bicycle during a sunny afternoon. She films herself in selfie mode with natural handheld movement, subtle hand shake, realistic smartphone autofocus, authentic stabilization, and true-to-life colors. Smiling at the camera, she says, "Let's explore Kyoto today!" She happily rides through narrow traditional streets lined with wooden houses, then notices she's taken a wrong turn and laughs, saying, "I think I'm lost!" Continuing down a peaceful path, she unexpectedly discovers a beautiful hidden Japanese shrine surrounded by tall trees and red torii gates. She parks her bicycle, walks toward the shrine with a surprised smile, captures a few quick shots of the peaceful scenery, then turns the camera back to herself and says, "Getting lost was worth it!" She gives a warm wave, smiles brightly, and says, "See you next time. Bye!" before reaching toward the phone to end the recording. The video should feel exactly like a genuine smartphone vlog with natural human movement, consistent facial features, realistic bicycle riding, authentic Kyoto atmosphere, no beauty filters, no CGI, no AI-plastic appearance, no subtitles, no logos, no watermark, and no background music—only birds, bicycle sounds, soft wind, footsteps, and natural outdoor ambience.

SCENE CONTEXT STORMWAKE. Mid-storm, high altitude, deep winter. A lone courier rides a four-winged stormvane at 180 km/h through a canyon of black thunderheads and ice-granite spires, and nearly collides head-on with an ancient Greatvane rising out of the fog. One unbroken 15-second oner, full third person. Rider and mount travel screen-left to screen-right for the first two thirds; the Greatvane enters head-on from screen-right depth. Camera flies as a second unseen flier, always outside the pair, never onboard. ACTIVE REFERENCES @kesh : mid-twenties courier, wiry, hunched low against the mount's neck, gloved hands locked on a harness strap, storm-glass mask over the lower face with two amber lamp-slits, wind-torn dark cloak streaming behind. 100% matches the reference. @stormvane: original species, not a dragon — serpentine flier, no forelimbs, two long lead wings plus two short rear stabilizer wings, charcoal-grey storm-scale over a keeled breastbone, ridged spine, fan-tipped whip tail, warm amber bioluminescence banded along the scale-seams, throat and wing-root channels. 100% matches the reference. @greatvane: the ancient one, same species grown enormous — slab-plated skull crusted with rime and lichen, wings scarred and translucent at the trailing edge, bio-seams burning deep ember instead of amber. 100% matches the reference. STYLE REFERENCE ONLY, not a fixed keyframe; model extends the world freely, never reproduced 1:1. LOCATION MAP Foreground: driving snow and sleet racing sideways across the lens at 90 km/h, occasional ice grit, torn cloud wisps passing at 20% frame coverage. Midground: @kesh and @stormvane, and later @greatvane, moving through a canyon of vertical ice-granite spires with wet black rock faces and hard snow shelves. Background: stacked black thunderheads, internal lightning firing every 2 to 3 seconds through the cloud canyons, distant snow peaks readable at 60% haze. Fog builds across the shot: 30% at the spires, 55% in the ravine, 70% at the fog wall the Greatvane breaks through. Camera stays on the shadow side of the pair, operator axis running low-left and slightly behind, rising to level as the shot opens out. SCALE LOCK @kesh stands one human tall in the saddle-harness. @stormvane measures wingtip to wingtip as wide as nine humans lying head to toe. @greatvane's skull alone is as long as twelve humans head to toe, its wingspan as wide as sixty; when its head fills frame, @kesh reads as a single fingernail against the eye socket. The size gap holds identical in every frame. FIRST FRAME / BLOCKING Non-empty opening frame: @stormvane is already mid-bank, left wing dropped forty degrees, @kesh already pressed flat to the neck with the cloak fully extended behind, snow already streaking past the lens, camera already moving at speed. Nothing is frozen, everything is already in motion. Rule of thirds: rider and mount held on the lower-left third, the nearest granite spire occupying the right third, sky and lightning filling the upper third. FORMAT MODE One continuous shot, one unbroken take, no cuts. The camera does not cut on its own. Speed changes are ramps inside the single take, not cuts. 0.0s to 7.0s — real time 7.0s to 7.5s — ramp down into extreme slow motion 7.5s to 11.5s — extreme slow motion, roughly one tenth speed 11.5s to 12.0s — ramp back up 12.0s to 15.0s — real time OPTICS 0.0s to 4.0s: wide chase at 84° FOV, camera low and tight behind the tail. 4.0s to 7.0s: reframe to 63° FOV as the camera arcs to the profile. 7.0s to 11.5s: settle to 29° FOV portrait compression on the Greatvane's skull and eye. 12.0s to 15.0s: open to 107° FOV super-wide rectilinear for the dive. Rectilinear with anamorphic optical flares, prime-lens character, 180° shutter motion blur, no drift mid-segment. CAMERA Flown as a second flier, matching @stormvane's 180 km/h with a half-beat lag on every direction change, 1 to 2 cm of buffet tremor from the crosswind, never mechanical. Height starts one wingspan below and behind the tail, rises to wing level on the arc, drops to eye level with the Greatvane's skull, then pulls back and up on the dive. Focus rides the rider's shoulders, pulls to the Greatvane's eye during the ramp, then racks to infinity as the ravine opens. Wide tonal latitude, soft highlight roll-off, highlights in the lightning hold detail rather than clipping. ACTION 0.0s to 4.0s — Subject: @stormvane banks hard between two ice-granite spires, wings carving the storm, wing membranes rippling and snapping taut, @kesh hunched low with the cloak lashing. Camera: chases low and tight behind, threading the same gap a half-beat later. 4.0s to 7.0s — Subject: @stormvane folds both lead wings and knifes into a narrow ravine between sheer cliffs, dropping 40 meters as it enters. Camera: arcs around to the side through a curtain of driving snow, lightning forking across the thunderheads behind the pair. 7.0s to 7.5s — Subject: @greatvane erupts head-on out of the fog wall directly ahead at 90 km/h closing speed. Camera: flinches back 2 meters and holds as time ramps into extreme slow motion. 7.5s to 11.5s — Subject: near-collision held at near standstill, the Greatvane's vast rime-crusted skull filling the frame within 4 meters, its enormous ember eye reflecting the tiny rider, breath steaming from the nostril slits in slow rolling plumes, snowflakes and ice shards suspended between the two creatures. Camera: drifts a slow 3 meters along the skull toward the eye, horizontal anamorphic flares raking the lens from the bio-seam glow. 11.5s to 15.0s — Subject: time cracks back to full speed, @stormvane rolls hard right under the Greatvane's jaw with 2 meters of clearance and dives down the ravine, the Greatvane banking away above into the storm. Camera: pulls back and up into a vast super-wide as the small flier plummets through swirling fog and lightning floods the peaks. PERFORMANCE @kesh reads as a working rider under strain, not a hero pose: shoulders locked, forearms corded on the harness strap, head turning three degrees to track the Greatvane, chin tucking at the last instant, chest visibly rising against the wind. Pore-level realism: vellus hair, asymmetric moles, capillary flush across the exposed cheekbone from cold, wet living eyes with catch-lights, visible breath and chest rise. Restrained and human, never cartoonish. Top-tier cinematic restraint. The Greatvane performs with age and indifference — the eye tracks the rider once, the pupil contracts, nothing more. PHYSICS Mass and inertia carry through every bank, fold and roll — @stormvane loads its wings before it turns and drifts wide out of it. Wing membranes flex and translate the load, they do not stay rigid. Snow and sleet obey the true wind vector across the lens and shear around both creatures. Ice grit strikes @kesh 's mask and deflects. The cloak drags behind with real fabric weight, never floating. In slow motion the suspended snowflakes still hold micro-drift and rotation. Contact between harness, strap, glove and scale reads with correct pressure and contact shadows. Nothing floats. LIGHTING WB locked 5600K cold storm daylight, dim and even, sun fully occluded by cloud. Key is a broad soft top-light bleeding through the thunderheads, camera on the shadow side of the pair. Lightning fires as a hard 2200K-neutral rim from behind screen-right every 2 to 3 seconds, edging wings, cloak and snow curtain for a fraction of a second. The only warm source in the world is the amber bio-seam glow at roughly 2000K along @stormvane's scale channels and throat, which spills onto @kesh 's glove, jaw and mask, and the deeper ember glow of @greatvane, which fills the slow-motion frame from below. Faint atmospheric haze throughout, thickening into the ravine. COLOR GRADE Cold steel-blue in the wet granite, slate-grey in the packed snow shelves, near-black in the thunderheads; every cold tone comes from the sky bouncing off rock and ice, not from a filter. Amber lives only where the bio-seams emit it — the wing-root channels, the throat, the reflected smear on wet leather and glove. The Greatvane's ember reads a full stop deeper and duller, an old fire against a young one. Blacks stay open with detail in the cloud, highlights roll off in the lightning. Color tied to source and surface, not a flat list. WARDROBE Layered fur-and-leather flight gear, storm-worn — cracked leather at the shoulders, salt-stiff fur at the collar, ice crusting the seams. Wind-torn dark cloak, hem frayed and snapping. Heavy grip gloves with a worn palm patch. Storm-glass mask over nose and mouth with two amber lamp-slits, fogging faintly at the edge from breath. Harness webbing tensioned across the mount's neck with visible strain on the straps. AUDIO NO MUSIC. SFX ONLY — diegetic sound and live audio throughout. No score, no soundtrack, no background music of any kind. Screaming 180 km/h wind, membrane wings cracking and snapping through the gusts, sleet peppering leather and glass, deep rolling thunder with 2-second delay, @kesh 's breath sharpening behind the mask. Audio ducks to a low pressurized hum and a single held breath during the slow motion, then the Greatvane's roar hits full and the wind returns. No dialogue, no subtitles. STYLE 8K photorealistic — no 3D render, no game engine, no game-cutscene aesthetic. Naturalistic master cinematography, fine film grain, 35mm film character, professional color grading, sharp focus with depth-of-field mastery, hyperdetailed scale, feather-scale and ice texture, steadicam fluidity carried into a flying rig. OUTPUT SETTINGS 15 seconds, one shot, 16:9, 8K ultra-high-definition, anamorphic flare character, real-time 0.0s to 7.0s and 12.0s to 15.0s, extreme slow motion 7.5s to 11.5s, ramps at 7.0s and 11.5s. POSITIVE LOCKS One unbroken take start to finish — the camera does not cut on its own, and every speed change is a ramp inside the same continuous move. @kesh and @stormvane stay identical every frame, no morphing, no drift, the mask, cloak, harness and bio-seam pattern consistent from first frame to last. @stormvane keeps four wings and no forelimbs throughout. The size relationship holds exactly: @greatvane's skull stays as long as twelve humans head to toe against a one-human rider. Camera stays outside the pair as a second flier and remains third person for the full 15 seconds. Kesh's eyes stay natural — no eye glow; the only emitted light in the world is the bio-seam glow on the two creatures. Contact shadows read clearly where glove meets strap and strap meets scale. Wings clear the granite with visible margin on every pass. Amber stays the single warm color in an otherwise cold frame. Snow travels on one consistent wind vector. Clean frame throughout — no logos, no UI, no text overlays, no subtitles, no watermark, no additional creatures or riders in shot.

Inside a traditional Japanese countryside farmhouse, a young Japanese woman and a young Japanese man turn an ordinary homemade dinner into a memory they'll never forget. She playfully steals the last bite of perfectly grilled river fish with her chopsticks, sparking a lighthearted chopstick battle filled with genuine laughter and warm chemistry. They continue sharing steaming rice, miso soup, seasonal vegetables, handmade pickles, and freshly brewed green tea while exchanging quiet conversations and comfortable silence. As the evening slows, they gently toast with handcrafted ceramic cups before sharing a single ripe strawberry, ending the night with peaceful smiles as rain softly falls outside. Shot with nostalgic handheld cinematography, warm lantern light, authentic Japanese farmhouse interiors, vintage Kodak Vision3 35mm film texture, subtle film grain, soft cinematic depth of field, realistic skin texture, natural micro-expressions, physically accurate body movement, gentle focus breathing, emotionally rich storytelling, and the quiet warmth of a timeless Japanese slice-of-life film. Every frame should feel like a genuine memory captured on film rather than a polished commercial, with ultra-photorealistic 8K HDR live-action quality and an intimate atmosphere where the smallest moments carry the deepest emotions.

15-sec Seedance prompt — 5 shots Character/style: Young Korean woman, 24, naturally attractive, realistic skin, long dark hair loosely tied back, minimal makeup. Loose cream cardigan, white T-shirt, blue jeans, and silver necklace. Cozy Seoul apartment, warm natural light. Preserve exact identity, outfit, hairstyle, and facial features across all shots. Authentic early-2000s MiniDV: 4:3, handheld shake, slight autofocus hunting, tape grain/compression, imperfect exposure, natural motion. No cinematic polish. SHOT 1 — 0:00–0:03 Wide MiniDV shot. She walks into frame, notices the camera, fixes her hair and cardigan, then looks into the lens with a shy smile and prepares to pose. SHOT 2 — 0:03–0:06 Cut closer. She poses deliberately for a photo, tilting her head and smiling sweetly. After a moment, she struggles to keep a straight face and looks away. SHOT 3 — 0:06–0:09 Different handheld angle. She tries to compose herself, presses her lips together, then accidentally lets out a tiny laugh. She looks down, embarrassed. SHOT 4 — 0:09–0:12 Cut back wider. She makes one final serious pose, holds it for half a second, then completely bursts out laughing. She covers her mouth and bends forward slightly. SHOT 5 — 0:12–0:15 Closer shot. Still laughing, she shakes her head and gives up posing. She smiles directly at the camera, walks toward it, and reaches for the lens. Her hand covers the camera. CUT TO BLACK at 15 seconds. Keep the laughter spontaneous and natural, with consistent character appearance throughout.

15 seconds, 16:9, 24fps, single-shot realistic sci-fi war short film. Image 1 is the sole reference for character, machinery, environment, composition, materials, and colors. Maintain the same adult female pilot, the same mud-stained white tech suit, the same purple-green bionic mechanical head, the same frozen wasteland, and the same overcast rainy weather. Do not alter the mechanical head's outline, fractured angles, cracks, armor positioning, exposed cables, or damage level. Start from the precise moment in Image 1. Mid-wide composition at ground level, the wrecked massive mechanical head dominating the left side of the frame, the adult female pilot kneeling in the mud below right, forehead pressed tight against the mechanical jaw, hands braced against the cold armor. The distant wasteland, debris, and low ridges remain blurred. The shot advances with an extremely slow, steady ground-level push toward her, no arcing, no rising, no cuts, no full heroic reveal of the mechanical head. 00:00 to 00:03.20: She keeps her forehead pressed to the armor, as if waiting for a response from within. Her breaths come short and irregular, shoulders jerking up suddenly with each inhale, then forcibly held down. Her right palm presses tight to the jaw seam, fingers gradually tightening, tips slipping faintly on the slick metal; her left hand slowly traces the crack, searching for any lingering vibration. Her knees sink deep into the mud, boot toes scraping backward from the strain, carving a short mud streak. Rainwater streams steadily down her white hair, lashes, the back of her hands, and the fractured purple armor. 00:03.20 to 00:06.20: She suddenly holds her breath, as if catching something faint. She lifts her head just a few centimeters from the armor, pressing her left ear to the mechanical face near the eye socket. Do not stand fully, do not turn toward the camera. Her eyes widen in terror, lips parting slightly, but the sound catches in her throat. A faint, intermittent dark orange electrical pulse flickers inside the mechanical eye, lasting less than half a second before dying out. A damaged relay's attempt to connect echoes from within, then falls silent completely. The mechanical body never revives, never turns its head, never blinks, never moves on its own. 00:06.20 to 00:09.30: She finally grasps that there is no response. Her jaw clenches first, nostrils flaring, tears mingling with the rain. She balls her right hand into a fist and strikes the mechanical jaw once, hard. The impact comes from a real shoulder-and-torso drive, the fist losing power immediately upon contact with the metal, sliding down the muddy armor. She does not pound repeatedly, does not exaggeratedly toss her head. She leans close to the mechanical face and speaks in the hoarse, breaking, barely coherent English of an adult woman, breath fractured almost beyond full respiration: [English, adult woman, hoarse, breath breaking] Open your eyes… please. Uncontrollable gasps punctuate each word, the final "please" whispered soft, no shouting, no theatrical sobbing. 00:09.30 to 00:12.60: Pause 0.4 seconds after the last word. No response stirs from the mechanical interior. She tries to inhale again, but her chest completes only half the motion, her throat emitting a short, stifled sound of failure. She grabs the armor edge with both hands, pulling her body closer as if her weight alone could force it awake. Her arms tense briefly, then the strength drains away entirely. Her spine folds inward, forehead slamming back against the mechanical jaw, shoulders beginning to tremble asymmetrically and violently. A low, uncontainable sob escapes her chest, but it forms no second line of dialogue. 00:12.60 to 00:15.00: She does not recover, does not look up. Her left hand clings desperately to the armor seam, while her right gradually loosens, fingers slackening one by one until the hand rests limp in the mud. The shot halts its advance a few paces from her, preserving the scale between her curled form and the vast dead machinery. In the distance, a piece of damaged metal debris shifts with the wind, scraping out a hollow friction sound once. The final second holds only the rain, her unsteady breathing that refuses to even out, and the blank absence of any mechanical hum from within. Performance requirements: Emotion progresses from denial and seeking signs of life, to fleeting hope, to confirming death, culminating in physical collapse of support. All emotions must manifest through breathing, swallowing, finger pressure, shoulder tremors, gaze, and shifts in center of gravity. No sustained screaming, no head-thrown-back wailing, no performing toward the camera, no rapid sobbing, no exaggerated slapping of the armor. Physics requirements: The mechanical head remains heavily embedded in the mud throughout, allowing only rainwater, loose cables, and minimal metallic settling to produce passive motion. The pilot's knees stay grounded at all times, her palms maintaining credible contact with the armor, no clipping or floating. The white tech suit clings to elbows and knees after absorbing water, mud realistically smearing with movements, without altering the garment's design. Hair is rain-flattened, moved only faintly by side winds. Sound: Close-up cold rain pattering on fractured ceramic armor, low-speed wasteland wind, wet fabric friction, glove slides on rough metal, one failed relay connection from the mechanical interior, one heavy but restrained fist impact. Dialogue centered, intimate range, retaining breaths and throat sounds. No music in the first 10 seconds. After confirming the machinery's total silence, introduce only a single extremely low, slowly evolving bowed string tone, volume below the rain, no piano, choir, trailer stabs, heroic motifs, or swelling sentiment. Prohibited: Mechanical revival, sustained eye glow, energy bursts, flashbacks, spectral apparitions, additional characters, enemy appearances, explosions, weapons, text, subtitles, camera shake, slow motion, rapid cuts, excessive depth-of-field shifts, garment deformation, mechanical structure redraws, weather changes, day-night alterations, or deviations from Image 1's cold gray low-saturation colors.

A continuous 15-second cinematic shot. An immortal swordmaster stands atop a misty cliff. He slowly draws a glowing blue ancient sword. The camera smoothly orbits around him as shimmering spirit particles swirl into the air. He steps forward, executes a fluid 3-step sword sequence, and thrusts his blade forward. A colossal spectral dragon formed of blue energy bursts from the sword tip, flying across the canyon and tearing through the clouds into the distant sunset.

Authentic late-1970s Mediterranean documentary captured on vintage 16mm Kodak film, naturally faded colors, subtle film grain, real optical imperfections, slight gate weave, shoulder-mounted handheld camera with imperfect human movement, soft vintage lenses, warm afternoon sunlight, realistic skin texture, no cinematic polish, feels like forgotten archival footage discovered decades later. 0–3 seconds: A close handheld tracking shot follows a young woman in her 20s walking slowly through a narrow seaside alley lined with whitewashed houses, hanging linen curtains gently moving in the ocean breeze. She wears a simple linen dress, a woven shoulder bag, and naturally messy hair with no makeup. Instead of posing, she lightly brushes her fingertips across the textured walls while quietly observing everyday life around her. 3–7 seconds: The camera naturally falls slightly behind her as she enters a tiny open courtyard where local people casually gather. An elderly man repairs a bicycle, someone waters colorful plants from a balcony above, laundry sways overhead, and a sleepy orange cat crosses directly in front of the lens. The camera subtly adjusts to avoid the cat, creating an authentic human reaction. 7–11 seconds: A young child runs toward a rolling wooden toy that passes close to her feet. She instinctively bends down, catches it with one hand, smiles warmly, and hands it back without stopping her walk. The child laughs and runs away. No one performs for the camera; everything feels naturally observed. 11–15 seconds: She reaches a small overlook facing the sparkling sea. Instead of stopping dramatically, she casually rests one elbow on an old stone railing while watching distant fishing boats. A gust of wind lifts loose strands of hair across her face. She gently moves them aside and continues walking out of frame as the handheld camera hesitates for a second before slowly following behind. Image Quality: Ultra-photorealistic vintage documentary, authentic analog exposure, realistic human motion, organic focus breathing, imperfect framing, Kodak 16mm archival texture, subtle light leaks, soft highlight bloom, natural skin pores, no AI smoothness, 4:3 aspect ratio.

Full magical transformation sequence: a moon knight floating in an ancient pale-blue parchment void as deep indigo ink spreads through the air like night dye blooming in water ,calligraphy strokes lash around the body, staining the hair into silver-white, sketching crescent armor seams, glowing cheek markings, and a curved moonblade silhouette in elegant blue brushwork. A circular ancient seal script sigil burns underfoot with splashes of silver light, the camera pushing in through drifting ink veils like pages turning in a storm, each brush impact synced to a rising drumbeat, ending in a blade-raised signature pose locked inside an indigo-and-silver calligraphy burst. Stock-footage grandeur, played sincere.

Gritty 16mm cinéma vérité, authentic 1970s Paris documentary footage, organic film grain, pure natural daylight, heavy shoulder-mounted handheld shake, no stabilization, soft focus falls, subtle light leaks, raw and elegant, lived-in realism. 0-3s: [Close Handheld Tracking] Young woman in her mid-20s, completely natural face with no makeup, soft features, wind-tousled hair, simple 1970s coat and scarf. She walks past a small floristería on a Paris side street. Camera stays tight on her profile and upper body as she slows for half a second, glances at the flowers, then looks toward camera with a quiet, unguarded expression. Instant natural beauty and presence. 3-7s: [Unsteady Following Shot] She continues into a lively sunlit plaza. Camera stays behind and slightly to the side, bouncing with every step. Children run and play all around her, pigeons scatter, people cross paths, soft Paris light and movement everywhere. Pure dynamism and street energy. 7-11s: [Handheld Medium] A group of kids chasing a ball suddenly cut right in front of her. One almost collides with her; she steps aside gracefully, laughs under her breath, and briefly exchanges a look with the child. The plaza stays alive and chaotic around them. 11-15s: [Raw Tracking] She keeps walking through the square, camera following close and imperfect, catching the residual energy of the kids, the light shifting across her face, and the natural rhythm of 1970s Paris street life without ever settling. Photorealistic 16mm film emulation, heavy organic grain, authentic period Paris details, natural skin texture, imperfect handheld physics, subtle film weave, coherent motion, pure cinéma vérité texture, no artifacts, 4:3 aspect ratio.