Community

A 15-second vertical stop-motion video of a cute clay toy miniature town growing over 50 years on one grassy diorama, told through five hard-cut year jumps (spring, summer, autumn, winter, sunset). Gentle, kid-friendly mood, warm palette. EXAGGERATED STOP-MOTION STYLE: animate as if photographed one frame at a time and played back at 12 frames per second, every pose held for two video frames (animation on twos). Movement is clearly choppy and snappy: characters and objects jump from pose to pose with visible gaps between poses rather than smooth interpolation. Push the handmade feel: strong squash and stretch on the clay builders, hammers and tools that pop and overshoot, small clay wobbles and jitter on every figure, slightly boiling fingerprint texture on the plasticine surfaces, tiny frame-to-frame shifts in the clouds and grass tufts, and subtle light flicker between frames. Crane arms and roofs move in distinct step-by-step jerks. The camera never glides smoothly: the final pull-back advances in visible steps, like a camera nudged between photographs. Matte plasticine with visible fingerprints. Tilt-shift miniature look: a sharp focus band across the middle of the diorama, soft blur at the top and bottom of the frame. Puffy cotton-like clouds in every shot. The diorama layout never changes: green hill on the left, blue river bend on the right, road through the center, circular town-square plot in the middle, toy crane in the back-right corner. The same rounded, chunky clay builders (oversized round heads, yellow hard hat, orange safety vest, blue overalls, brown boots, toy hammer and paint roller) appear in every shot. Do not render any text, numbers, captions or signs in the picture. No music, no speech, no sound. SHOT 1 [0:00-0:03] - Year 1, spring, empty meadow. Wide view of the whole diorama, tilt-shift, focus on the middle band where the town-square plot sits. High three-quarter angle looking down, diorama centered. Locked-off camera, no pan, tilt, push, pull, orbit, drift or zoom. A clay builder taps a small stake into the ground with big snappy squash-and-stretch hammer hits and plants a tiny flag in the town-square plot. Ends on the flag planted. No buildings yet beyond the crane. CUT TO SHOT 2 [0:03-0:06] - Year 10, summer, building begins. Same vantage, slightly closer, same focus band, diorama centered. Locked-off. Scaffolding stands around several half-built small houses; several builders hammer with exaggerated bouncy pops; the crane lowers a roof onto a house in jerky steps. Ends on the roof settling onto the house. Bright green grass, sunny light. CUT TO SHOT 3 [0:06-0:09] - Year 25, autumn, the town takes shape. Closer view of the main street and town square, same focus band, centered on the road and square. Locked-off. Houses, a shop and a school are painted in bright colors; trees have orange autumn leaves; builders roll paint and carry bricks with hopping, bouncy steps; the crane lifts the top piece of a clock tower. Ends with the clock tower nearly complete. CUT TO SHOT 4 [0:09-0:12] - Year 40, winter, the town is nearly finished. Wide view of the whole diorama, same focus band, same high three-quarter angle as shot 1. Locked-off. Light snow dusts the roofs and grass; small warm lights glow in the windows; builders take down the scaffolding piece by piece; the town square is finished. Ends with the clock tower free of scaffolding. CUT TO SHOT 5 [0:12-0:15] - Year 50, the finished town glows at sunset. Starts on the town square and clock tower, ends on the whole diorama in a wide view; focus stays on the middle band. One slow pull-back advancing in visible stepped jumps away from the town square until the whole diorama fills the frame; builders wave with big bouncy arm swings. Ends on the whole finished town, builders waving, in a hold of about one second. Warm golden sunset, glowing windows, puffy pink-gold clouds. No orbit, no tilt, no zoom, no cut.

STYLE: a real British independent feature, shot on 35mm Kodak Vision3 film by Robbie Ryan for Andrea Arnold and by Gregory Oke for Aftersun. Naturalistic available light, real London locations, nothing staged or polished, the camera observes a moment that was already happening. Muted true colours, soft contrast, slightly lifted blacks, visible photochemical grain, gentle halation, real lens imperfections, focus that is not perfect everywhere. Faces with real skin texture, never smoothed. CASTING: @asset is JOY, 12: she is exactly that girl, same face, same natural hair in two high puffs with thin red ties, same clean red football shirt with a white collar and no logo. Take ONLY her identity and wardrobe; ignore its framing and its light. MOMENT: early morning at low tide on the Thames foreshore, grey and cold. Joy juggles a scuffed football on her knee, counting under her breath, totally absorbed; the ball is in the air. FRAMING: wide 21:9, she is small on the left third, full body, standing on wet shingle and mud; the river is a flat pale grey; across the water, far away and soft in the morning mist, the silhouette of Tower Bridge. FOREGROUND: an old wooden mooring post and a coil of rope, out of focus, in the lower right. LENS: 50mm at 1.2 m. LIGHT: flat overcast dawn, a slightly brighter patch of sky behind the bridge; no golden glow. RULES: she is dignified and determined, clean and well cared for. Nobody looks into the lens. Correct anatomy, natural hands with five fingers. No text, no logos, no numbers. It must look like a frame from a real film, never like an illustration, a render or an advert.

A poster for an independent auteur ANIMATED short film for young adults, tender and funny with real loss underneath, the kind that wins at Annecy or Ottawa, with the attitude of an A24 film, designed by a great graphic designer. ONE single impossible image treated with complete seriousness and ONE bold dominant colour that covers most of the poster as a flat colour field. The whole poster is a 2D ANIMATION still made in two-colour risograph in lilac and tomato red, grainy, with misregistration; the hand-made texture is clearly visible. Characters with an auteur design: odd proportions, long noses and long necks, dot eyes, asymmetry, never big eyes, never cute, never anime or Pixar. IMAGE: a flat lilac bedroom floor seen from above: two teenage best friends with long necks lie head to head inside one enormous red sleeping bag that stretches across the whole floor like a road, whispering, on their last night before one of them moves to another country. Simple bold composition, lots of empty space. TYPOGRAPHY: at the top, small spaced-out capitals: "A FILM BY ARANTXA BARCIA". The title hand-lettered in the same style: "SLEEPOVER". At the bottom, small: "OTTAWA INTERNATIONAL ANIMATION FESTIVAL" and a line of small credits. No brand names, no signs, no logos, no readable writing in the image. All text perfectly spelled, crisp and legible, with a generous margin from every edge of the poster. Lots of negative space.

FORMAT: 21:9 anamorphic widescreen, 30 seconds, every composition designed for the wide frame. · GENRE AND TONE: a scene from OBĀCHAN, a Japanese live-action short film selected at Short Shorts Tokyo: deadpan, tender and funny, about a tiny grandmother who raised her enormous grandson and dresses him for his first sumo bout. NARRATIVE AND SUBTEXT (the whole story must read from the images alone): KENTA, 19, a huge young sumo wrestler, wakes up nervous on the day of his first official bout; his tiny GRANDMOTHER, 85, who has raised him alone, is already up. She cooks him an enormous breakfast, and he eats it all. She makes him do his leg-raising exercises in the tiny kitchen: each time his foot stamps down, all the cups jump. At the stadium, in the narrow dressing room, she stands on a wooden step to tie his white belt herself, pulling with all her small weight; he holds his breath. He walks out to the ring; she sits alone in the stands, very small, holding his towel. THE BOUT: he pushes his opponent out in one second. THE END: in the corridor afterwards, he lifts his tiny grandmother and sets her on his shoulder; she sits there very seriously, holding on to his hair knot. REFERENCE: Source A is the film poster: it defines ONLY the look and the two characters: a flat salmon-pink backdrop, soft side light, the tiny grandmother with white hair in a grey kimono, and KENTA, a huge young sumo wrestler with a black topknot and a white belt. IGNORE ALL THE TEXT AND TYPOGRAPHY OF THE POSTER: there is no text in this video. It does NOT define the framing: this film is in 21:9. HARD RULES, for every shot: 0. DEPTH: every shot has three layers: an object close to the lens, soft and out of focus; the action in the middle; a rich background; never a flat frame with nothing between the lens and the characters. 1. NO TEXT, NO LETTERS, NO LOGOS anywhere, no writing on walls or banners. 2. PALETTE exactly like the poster: a flat saturated salmon pink (#E9B3A8) fills most of every frame (the flat walls, the dressing room, the stands), her grey kimono, his skin, one white for the belt; real 35mm film texture and grain, true colours, never brown, never yellow cast. 3. LIGHT: soft but directional side light like an Irving Penn portrait, crisp soft shadows, never flat. 4. KENTA is three times her size; she reaches his chest only when she stands on a step. 5. Only two sumo wrestlers in the bout; the crowd is soft and far. SHOT 1 (0.0 - 2.0 s): EXTREME CLOSE-UP, macro: FOREGROUND (soft, close to the lens): a steaming rice bowl. a huge eye opens in the morning light, a strand of black hair across it. CAMERA: slow push-in. · SHOT 2 (2.0 - 4.0 s): TOP SHOT straight down on a tiny salmon-pink kitchen: FOREGROUND (soft, close to the lens): a paper lantern. the tiny grandmother at the stove fills a huge bowl; Kenta fills half the room at the small table. CAMERA: slow rotation above. · SHOT 3 (4.0 - 5.5 s): MACRO: FOREGROUND (soft, close to the lens): the edge of the rice pot. grains of rice tumble from a wooden spoon into an enormous bowl. CAMERA: slow tilt down with them. · SHOT 4 (5.5 - 7.0 s): CLOSE-UP, his face cut by the top edge, only his cheeks and mouth: FOREGROUND (soft, close to the lens): a teapot. he eats, very seriously. CAMERA: slow lateral glide. · SHOT 5 (7.0 - 9.0 s): WIDE, very low at floor level, 14 mm: FOREGROUND (soft, close to the lens): a stack of cups. Kenta raises one huge leg high in the tiny kitchen and stamps it down; on the table all the cups jump; the grandmother claps once. CAMERA: slow push-in at floor level. · SHOT 6 (9.0 - 10.5 s): INSERT: FOREGROUND (soft, close to the lens): a chopstick holder. cups and chopsticks jumping on the table as the floor shakes. CAMERA: slow lateral glide. · SHOT 7 (10.5 - 12.5 s): EXTREME WIDE from very far down a long salmon-pink corridor of the stadium: FOREGROUND (soft, close to the lens): a corridor pillar. the huge young man and the tiny old woman walk side by side away from us. CAMERA: slow following dolly. · SHOT 8 (12.5 - 14.5 s): MEDIUM in the narrow pink dressing room: FOREGROUND (soft, close to the lens): a hanging white towel. she climbs onto a small wooden step and wraps the long white belt around his belly, pulling with her whole body; he holds his breath. CAMERA: slow arc around them. · SHOT 9 (14.5 - 15.5 s): MACRO: FOREGROUND (soft, close to the lens): a wooden step. her small wrinkled hands pull the knot of the white belt tight. CAMERA: slow push-in. · SHOT 10 (15.5 - 17.0 s): CLOSE-UP of Kenta from below, his face high and serious, her white hair at the bottom edge. CAMERA: FOREGROUND (soft, close to the lens): a curtain of the dressing room. slow tilt up. · SHOT 11 (17.0 - 19.0 s): EXTREME WIDE TOP SHOT straight down on the round clay sumo ring: FOREGROUND (soft, close to the lens): the purple tassel of the ring roof. two wrestlers crouch face to face, tiny in the middle of the frame. CAMERA: slow rotation. · SHOT 12 (19.0 - 20.5 s): WIDE, long lens: FOREGROUND (soft, close to the lens): a spectator's shoulder. in the soft stands, one tiny old woman sits alone holding a folded towel on her lap, perfectly upright. CAMERA: slow push-in. · SHOT 13 (20.5 - 22.5 s): MEDIUM, low at the edge of the ring: FOREGROUND (soft, close to the lens): the straw edge of the ring. the two wrestlers clash; in one second Kenta pushes the other out over the straw edge. CAMERA: fast lateral tracking with the push. · SHOT 14 (22.5 - 24.0 s): MACRO of the grandmother's hands on the towel: FOREGROUND (soft, close to the lens): the folded towel. they squeeze it once. CAMERA: slow tilt up to her face, which does not move at all. · SHOT 15 (24.0 - 26.5 s): MEDIUM in the pink corridor: FOREGROUND (soft, close to the lens): a corridor doorframe. Kenta lifts his grandmother and sets her on his shoulder; she sits very seriously, holding his topknot. CAMERA: slow rise. · SHOT 16 (26.5 - 30.0 s): EXTREME WIDE from the end of the long pink corridor: FOREGROUND (soft, close to the lens): a hanging banner rope. the huge young man walks away with the tiny old woman sitting on his shoulder. CAMERA: very slow following dolly. · CINEMATOGRAPHY RULES, for every shot: an auteur's original visual language, every frame a beautiful, surprising composition: NEVER a frontal postcard, the subject off-centre or cut by the edge of the frame; EVERY shot has real depth: an out-of-focus foreground element filling 20-40 % of the frame, the action in the middle ground and a rich background; great variety of angles from shot to shot, as written in each shot: top shots straight down, worm's-eye nadir shots from the ground looking up, high angles, low angles, extreme wide angles with a 14 mm lens close to the ground with strong perspective, long lenses compressing the background, inserts and big close-ups; never two consecutive shots of the same size or the same height. FRAMING VARIETY: the framings change radically from shot to shot: real MACRO shots (an eye, a fingertip, a crumb, a thread filling the frame), extreme wides where the character is tiny in a huge field of colour, strange off-balance compositions with the subject cut by the edge or placed in a corner, very low at ground level, straight down from above, from behind, from far away through the space; each framing feels unexpected and beautiful. CAMERA MOVEMENT RULE: EVERY SHOT HAS ITS OWN CAMERA MOVEMENT, written in the shot, already in motion from the first frame, smooth and motivated by the character; the move is defined by what enters or leaves the frame; never a zoom; hard cuts on movement. STORY RULE: every action is big, clear and readable at first sight; the whole story must be understood from the images alone, with no words and no text. ACTING RULES: real, specific, imperfect human behaviour, never posed: each character wants something and something stands in the way; show it with small physical behaviour: breathing, a hesitation, a gesture started and abandoned, eyes that look and look away; nobody holds a pose more than half a second; nobody looks into the lens; every face is natural and real, with real skin texture, resembling no actor; faces are either close enough to read or seen from behind or in silhouette, never small deformed faces in the background. · AUDIO: ABSOLUTELY NO MUSIC, NO SCORE, NO SINGING, NO VOICES, NO DIALOGUE. Only real sounds of the place, described in each film. · Sounds: a morning kitchen, rice pouring, a huge foot stamping and cups rattling, footsteps in a corridor, a belt pulled tight, a distant crowd, the slap of two bodies clashing, a short roar of the crowd. · DO NOT: no text, no banners, no music, no other grandmother, no injury, no overcast, no brown cast.

FORMAT: 21:9 anamorphic widescreen, every composition designed for the wide frame. · GENRE AND TONE: LA FERIA, an EXPERIMENTAL black-and-white visual short film in the tradition of the great Spanish street photographers of the 1950s and 60s and of poetic cinema (Víctor Erice, Sergei Parajanov, Chris Marker). There is no plot to explain: it is a sequence of strong, strange, beautiful images with one quiet idea, one visual rhyme and a final image that surprises. · CONCEPT: dawn after the village feria in Andalusia: everything is ending. A young woman in a polka-dot flamenco dress sleeps in the back of a horse cart as an old man leads it home along the dusty road. Images of the end of a party: lanterns coming down, a lost shoe, the horse's eye, polka dots, dust. The visual rhyme: the flower in her hair falls onto the road, and the last image is the empty road full of flowers that fell from every cart. · REFERENCE: @asset is the film poster: it defines ONLY the look, the world and the people. IGNORE ALL THE TEXT AND TYPOGRAPHY OF THE POSTER. It does NOT define the framing: this film is in 21:9. · HARD RULES, for every shot: 1. BLACK AND WHITE ONLY, Kodak Tri-X 400 film look, heavy grain, hard contrast, deep crushed blacks and bright whites; if any colour appears, the shot has failed. 2. NO TEXT, NO LETTERS anywhere. 3. Every shot is a radical, original composition, never centred, never symmetrical, never a postcard: ultra-wide lenses very close to faces and objects, extreme low or high angles, frames inside frames (doorways, mirrors, windows) with half of the image in pure black, tilted horizons. 4. Every shot has deep layers: something very close and huge in the foreground, the subject in the middle, a world behind. 5. Light is hard and graphic: blades of sun, stripes through shutters, silhouettes, faces half in darkness. 6. The people are real and unposed, caught in the middle of what they do, never posing, never looking into the lens, every one moving in their own way. 7. Each shot holds and breathes; slow, minimal, precise camera movement; hard cuts. · SHOT 1 (0.0 - 3.5 s): Ultra-wide, the lens almost touching her: the face of the young woman asleep against the wooden side of the cart fills the bottom left of the frame, mouth open, the flower in her hair; behind her, small and sharp, the old man leading the horse along the dusty road under a white sky. · SHOT 2 (3.5 - 7.0 s): Low angle from the dust of the road, a huge wooden cart wheel turning in the foreground, cut by the frame: the cart passes, the horse's legs, the old man's boots. · SHOT 3 (7.0 - 10.5 s): Extreme close-up of the horse's eye, with lashes, reflecting the bright morning road and the tiny shape of the cart. · SHOT 4 (10.5 - 14.0 s): Through the skeleton of a dismantled feria booth, its bare wooden poles making black lines across the frame: men take down strings of paper lanterns against the white sky; the lanterns fall slowly. · SHOT 5 (14.0 - 17.5 s): Top shot, straight down: one abandoned flamenco shoe lying in the dust, the shadow of the passing cart sliding over it. · SHOT 6 (17.5 - 21.5 s): Tilted close-up from the side of the cart, the polka-dot ruffles of her dress filling the foreground, bouncing with the road: beyond them, the old man walks without looking back. · SHOT 7 (21.5 - 25.5 s): Macro on the road: the flower slips from her hair and falls into the dust; the wheel passes beside it. · SHOT 8 (25.5 - 30.0 s): Extreme wide, low, from the middle of the road: the cart small and far away on the horizon, and the whole dusty road in front of the lens covered with fallen flowers from all the carts of the night. Hold. · AUDIO: SILENT. No audio at all. · DO NOT: no colour, no text, no music, no centred frontal shots, no symmetry, no posing, no smiling into the lens, no modern objects.

A minimalist, arty film poster for an independent short film, like an A24 poster designed by a great graphic designer. IMAGE: a photograph on film of a small pastel pink vintage ice cream van parked alone on a green clifftop in Wales on a grey day; the ice cream man, an old man in a white jacket, leans out of the serving window looking at the grey sea; no customers anywhere. Pale grey sky, green grass, one pink. TYPOGRAPHY: the title in a large, playful rounded serif across the top: "OUT OF SEASON". Below it, small: "a short film by Rhys Morgan". At the bottom: "BAFTA · BEST BRITISH SHORT FILM" and a line of small credits. No brand names, no signs, no logos in the image. All text perfectly spelled, crisp and legible. Lots of negative sky.

GENRE AND TONE: a scene from LITTLE ORBIT, a Mexican short film that won at Tribeca, shot on 35mm like Roma and the films of Fernando Eimbcke. Tender, funny, quietly emotional. Nothing is sentimental; it is told through objects and looks. · NARRATIVE AND SUBTEXT: TOÑO, 8, waits every night on the rooftop of his building in Mexico City, wearing the astronaut helmet he made from a cardboard box, because his father, who works far away, promised they would look at the moon at the same time. Tonight his father's phone call does not come. He keeps waiting. Then the rooftop door opens: his father has come home. · REFERENCE: @asset is the film poster: it defines ONLY the look, the world and TOÑO: a flat rooftop at blue hour, a huge full moon, laundry lines, plain black water tanks, a plastic white chair, the city lights below, one warm lit window; TOÑO, 8, in a navy sweatshirt, grey trousers and sneakers, wearing a cardboard-box helmet with a silver foil visor and a hand-drawn planet. IGNORE ALL THE TEXT, TYPOGRAPHY AND BRAND NAMES OF THE POSTER: the water tanks are plain black with no writing. It defines the composition of SHOT 1 ONLY. · HARD RULES, for every shot: 1. NO TEXT, NO LOGOS, NO BRAND NAMES, NO WRITING anywhere, not on the water tanks, not on the phone. 2. TOÑO is always the same boy with the cardboard helmet until SHOT 6. 3. THE FATHER: a Mexican man of 35, short black hair, a stubble beard, a worn denim work jacket, a dark canvas travel bag. 4. Deep blue dusk, the warm moon and one warm window as the only warm lights; 35mm grain, true colours. · SHOT 1 (0.0 - 4.0 s): WIDE matching the composition of @asset: Toño on the plastic chair, the moon rising over the city, the laundry moving in the wind. Very slow push in. · SHOT 2 (4.0 - 7.5 s): INSERT in his lap: an old cracked mobile phone with a dark screen; his small thumb taps it twice; nothing happens. · SHOT 3 (7.5 - 11.0 s): CLOSE-UP through the foil visor, 85mm: his eyes, the moon reflected in the foil. TOÑO (a whisper, in Spanish): "Houston... ¿me oyes?" · SHOT 4 (11.0 - 15.0 s): WIDE from behind: he stands up on the chair and stretches one arm towards the moon as if he could touch it; the city hums below. · SHOT 5 (15.0 - 18.5 s): MEDIUM, behind him: the metal door of the rooftop stairwell creaks open, warm light spills onto the wet concrete; the silhouette of the father with his travel bag. Toño does not notice. · SHOT 6 (18.5 - 22.5 s): MEDIUM CLOSE-UP of Toño: he hears it, turns slowly, the visor catches the warm door light; he lifts the box off his head: his face, mouth open, eyes wide. · SHOT 7 (22.5 - 26.5 s): MEDIUM: the father drops the bag and kneels; Toño jumps off the chair into his arms; the cardboard helmet rolls away on the concrete. · SHOT 8 (26.5 - 30.0 s): WIDE, the same angle as SHOT 1: father and son sit together on the plastic chair, the boy on his lap; the father is now wearing the small cardboard helmet, too small for him; both look at the moon. FATHER (dry, in Spanish): "Aquí Houston." Hold. · CAMERA RULE: one slow movement per shot or locked-off; hard cuts on settled frames. · AUDIO: the distant hum of the city, a far-away dog, laundry flapping, the creak of the metal door, the scrape of the cardboard on concrete. Each spoken line exactly once. A sparse score enters at 15.0 s: a soft nylon-string guitar and a piano, swelling gently at the hug and resolving on the last shot. No singing. · DO NOT: no text or logos on anything, no other people, no daylight, no phone screen lighting up with words, no crying faces, no orange grade.

GENRE AND TONE: a scene from FIRST SNOW, a Japanese short film awarded at the Berlinale Generation Kplus, shot on 35mm like the films of Hirokazu Kore-eda. A gentle, funny, luminous family film about two sisters. Nothing is sentimental; children behave like real children. · NARRATIVE AND SUBTEXT: HANA, 8, and her little sister MEI, 5, have just moved to Hokkaido from the south. Mei has never seen snow and is afraid of it. Hana, a bit bossy, wants her to love it. This morning the first snow falls, and Hana shows her how, by falling backwards into it first. · REFERENCE: @asset is the film poster: it defines ONLY the look of the film and the two girls: an immense field of fresh white snow under a soft overcast sky, only white and red; HANA, 8, black bob haircut, a red wool duffle coat with toggles, a cream scarf, black snow boots; MEI, 5, a red wool duffle coat, a grey knitted bobble hat and black snow boots. IGNORE ALL THE TEXT AND TYPOGRAPHY OF THE POSTER: there is no text in this video. It defines the composition of SHOT 8 ONLY. · HARD RULES, for every shot: 1. NO TEXT, NO LETTERS anywhere. 2. Only TWO children exist, always the girls of @asset: Hana with the black bob and no hat, Mei with the grey bobble hat. If a third child appears, THE SHOT HAS FAILED. 3. Soft white overcast morning light, snow falling gently in every exterior shot, true clean whites, the red coats the only strong colour. 4. 35mm film, fine grain, natural real faces. · SHOT 1 (0.0 - 4.0 s): INSERT inside a small wooden house: a frosted window at dawn; Mei's small hand presses on the glass; outside the first big flakes fall over a white field. · SHOT 2 (4.0 - 8.0 s): WIDE from the field towards the house: Hana bursts out of the front door and runs into the deep snow; Mei stops on the wooden doorstep and does not step down. · SHOT 3 (8.0 - 11.5 s): MEDIUM CLOSE-UP of Mei on the doorstep, 85mm: her lower lip pushed out; a snowflake lands on her coat and she brushes it off fast, as if it were a bug. · SHOT 4 (11.5 - 15.0 s): MEDIUM of Hana in the field, facing her sister: HANA (shouting, in Japanese): "Mei! Mite!" She spreads her arms and falls straight backwards into the snow, disappearing into it with a soft thump. · SHOT 5 (15.0 - 18.5 s): CLOSE-UP of Mei: a gasp, then a giggle escapes; she takes one step off the doorstep and her boot sinks into the snow up to her knee. · SHOT 6 (18.5 - 23.0 s): LOW TRACKING SHOT beside her: Mei runs clumsily through the deep snow towards her sister, falls forward, sits up with snow on her face and bursts out laughing. · SHOT 7 (23.0 - 26.5 s): CLOSE-UP from above of the two girls lying side by side, faces up: snowflakes fall onto their cheeks; Mei sticks out her tongue to catch one; Hana copies her. · SHOT 8 (26.5 - 30.0 s): TOP SHOT, perfectly vertical from high above, matching the composition of @asset: the two girls make snow angels in the immense white field, small in the centre; the camera rises slowly. · CAMERA RULE: gentle observational camera, one movement per shot, hard cuts on movement; never two consecutive shots of the same size. · AUDIO: the hush of falling snow, boots crunching, the soft thump of the fall, the girls' breathing and real children's laughter. The one shouted line exactly once. A light score enters at 15.0 s: a solo piano melody with soft pizzicato strings, playful, resolving warmly on the top shot. No singing, no adult voices. · DO NOT: no third child, no adults, no text, no blue tint, no sun, no snowball fight, no crying for real.

Subject Lili, a young woman with a reddish-blonde chin-length bob, remains the same person throughout: identical face, hair, body shape and identity in every outfit. Match her identity and opening everyday outfit to the lili references. Her normal outfit is the textured tan bucket hat, large round orange-tinted sunglasses, ribbed beige turtleneck sweater, rust-orange corduroy midi skirt and white canvas high-top sneakers. Costume changes follow the supplied designs: the mummy costume references for the mummy, the witch costume reference for the witch, and the female werewolf costume reference for the werewolf. Action A single uninterrupted 15-second camera take, one shot only, framed in 4:3. 0-3 s, OPENING: at the top of a steep snowy mountain run lined with pine trees, Lili stands still on her skis with ski poles in hand, in her normal everyday outfit. The camera is directly behind her, at eye level, in a steady, still medium shot, and she is centered in the 4:3 frame. Her back and the back of her hat and skirt face the camera and her skis point down the slope away from the camera. Her torso and shoulders are twisted around toward the camera behind her, so she looks back over her shoulder straight into the lens and smiles warmly. Then she turns her body forward to face the slope, bends her knees and pushes off down the run. 3-5 s: she launches downhill at full speed while the same physical camera smoothly lifts up and follows, like a drone taking flight, evolving continuously from the medium shot behind her into a wider aerial chase view with no edit. 5-12.5 s: the drone continues in one fluid motion, orbiting Lili in wide sweeping arcs from behind to her side to the front and back again, diving low, rising high and banking between the pines, always keeping her in frame with the mountain slope behind her. Maintain a heroic sense of speed with powder spraying from her skis. She performs natural athletic freestyle tricks in this order: a hard carve into a big jump with a spinning grab; a clean landing absorbed by her knees; a sideways power slide; then a huge jump with a full spin and a clean landing before racing onward. Wardrobe changes occur inside the same continuous camera movement, never as an edit or visual cut. Each transition happens as the moving drone passes a large dark pine trunk that briefly crosses the lens; the camera keeps moving through the occlusion with continuous time and space, and when the trunk clears Lili is already in the next outfit, with the same body and motion on the same downhill line: - About 6 s: mummy costume, aged cream-and-brown linen bandages wrapped around her body and arms, loose strips streaming in the air, with her face and hair free. - About 8.5 s: witch costume, black pointed hat with purple band, long black lace dress with wide sleeves, short cape and silver rings. - About 11 s: female werewolf costume, wolf-ear hood, grey fur cape, torn red-and-black plaid shirt, dark distressed denim miniskirt, torn black tights and fur boots. 12.5-15 s: still in the werewolf costume, the drone gradually decelerates and glides to a complete stop high on the slope in a wide shot, then remains perfectly still with no movement at all. Lili is seen from behind skiing away down the run, becoming smaller and smaller toward the valley until the final frame. The closing view is reached by the same continuous camera move, with no cut into it. Environment A steep snowy mountain ski run lined with pine trees; the same location remains continuous from the first frame to the last, with no scene change. Camera One physical camera, one continuous take, uninterrupted time and space from first frame to last. No cuts, jump cuts, dissolves, fades, whip-cuts, edited angle changes or second camera angle. The opening is a steady, still medium shot from behind her at eye level. The camera then lifts into a following drone move through wide arcs, low dives, high rises and banks between the pines, decelerates during 12.5-15 s and settles into a completely locked, motionless wide shot. Style Retro 1970s horror cinema: 35 mm film grain, saturated telecine color, slightly faded blacks, soft vintage contrast and cold overcast winter daylight on snow. Realistic skiing physics, natural balanced body posture, stable identity and continuous wardrobe. Audio Sounds only: rushing wind; skis carving and hissing on snow; snow spray; a soft whoosh each time a pine trunk passes across the lens; impact thuds on landings; calm wind and receding ski sounds during the final still hold. No music, score, singing, voice or dialogue. Constraints The entire video is one unbroken camera take with no cut of any kind. Lili's face, hair and body shape never change. No captions, subtitles, titles, labels, annotations or watermarks.

One continuous take, no cuts. A photoreal fashion film shot on 35mm: warm amber tones, film grain, shallow macro focus. The world freezes and only the camera moves. 0-2s: Camera starts low on his shoes on a sunny city street and cranes smoothly upward to reveal a man in a detective hat, a 90s Motorola brick phone held to his ear, an energy drink cup in his other hand. The supplied Kawi Camping logo sits on the detective hat, the cup and the phone. 2-4s: The camera tracks backward ahead of him at his walking pace as a passer-by walking an alligator on a leash comes the other way. 4-5s: Their elbows clip. He falls back, letting go of the energy drink and the brick phone, and the detective hat flies off his face. The cup tips and one edge of its lid pops up at the top. 5-6s: Time almost stops. The camera widens to show the frozen accident, every object hanging in the air. 6-7s: Bullet time: the camera travels to the detective hat and orbits it and the nearby objects in macro, while the action stays almost frozen. 7-8s: The camera moves on to the tilted energy drink cup. A thin stream of energy drink escapes from the top through the gap where the lid has popped up. The bottom of the cup stays sealed and dry. 8-10s: The camera moves on to the open Motorola brick phone, its screen reading "BOSS" in clear legible text. 10-12s: The camera pulls back to the whole scene, then time snaps back: he lands and everything falls where its own path takes it. Sound: sunny city street ambience and the muffled, indistinct sound of his phone call (no intelligible words). When time freezes at about 5s the street sound and the call drop away to a low hum that holds through the bullet time. At about 10-12s, as time snaps back, the hum releases into the clatter of everything landing. No music. Branding: reproduce the supplied Kawi Camping logo exactly as in the reference (slanted black bold "Kawi" wordmark with "Camping" beneath it) on the detective hat, the cup and the phone. No other readable text apart from "BOSS" on the phone screen.

One continuous shot, no cuts. Extreme close-up of a young man's brow, eyes and bridge of the nose seen as a reflection in a narrow horizontal strip of glass, a car rearview mirror, framed above and below by dark metallic borders. He has short blond hair with straight bangs falling over the forehead, fair skin and light-coloured eyes, matching the person in the reference. Eyes centered in the strip at eye level, sharp focus on the eyes and brow, warm catchlights in the irises. Outside the glass, soft out-of-focus green foliage and gentle daylight. Timing: 0.0 to 1.0 s he stares intently and steadily into the mirror, vigilant and tense. At about 1.0 to 1.25 s a subtle gaze shift: the eyelids narrow slightly, the gaze drops downward and slightly away, with a barely perceptible tilt of the head downward. From about 1.25 s to the end of the clip he holds this lowered, off-axis line of sight, still and tense, conveying hesitation and inner tension. Camera: locked-off camera on the mirror. The subject moves inside the frame. No pan, tilt, push, pull, orbit, drift, zoom, or reframing. End view is the same framing with the gaze lowered, held. Look: soft naturalistic daylight with low-key car-interior shadows, warm skin tones, rich contrast, heavy 35mm film grain, vintage 1990s analog thriller look. Silent, no dialogue, no music. Show only the reflection in the mirror strip; do not reproduce the reference's layout, panels, white background, or full-body views, and add no text.

Four friends, Dancer 1, Dancer 2, Dancer 3 and Dancer 4, dance together on a New York City sidewalk on a fall day. Each dancer's face, hair, skin tone and body must match their own identity reference exactly, with no blending between people: the first four reference images are the only source of faces and hair, and the last four images show clothing only (their faces are not used). Dancer 1: man with a full rounded black afro, small stud earring, large silver ring. Dancer 2: lean athletic man with short dark-brown hair and a side-swept fringe, light grey-green eyes, angular jaw, tanned skin; his hair is completely dry, soft and matte, with no wet look, no water, no sheen, and he wears no necklace. Dancer 3: woman with light-brown hair slicked straight back with a side part (not curly), strong dark brows, grey-green eyes, laughing as she dances. Dancer 4: woman with dark-brown hair in a loose messy updo, warm tan skin, curvy build, layered gold necklaces with a nameplate pendant. Wardrobe (clothing only from the last four images): Dancer 1 in the brown glen-plaid double-breasted suit, light blue shirt and patterned tie. Dancer 2 in the olive/tan windowpane double-breasted suit, lavender shirt and tie. Dancer 3 in the newsboy cap, grey houndstooth double-breasted blazer, grey turtleneck, houndstooth trousers and tan knee boots. Dancer 4 in the hooded fair-isle knit, grey-brown houndstooth long coat and grey wool trousers. No handbags, folios or held props; hands free for dancing. Setting: a wide Manhattan sidewalk lined with brownstones and storefronts on a bright autumn day, yellow cabs softly blurred in the distance, orange, red and golden trees, fallen leaves on the pavement, crisp soft daylight. Only a few blurred pedestrians far in the background. Exactly four dancers in the foreground. Shots, with hard cuts between them and camera movement only inside a shot: Shot 1 (0-4s): locked-off medium-wide knee-up shot at eye level, the four dancers in a loose row facing camera, all four faces clearly visible, dancing joyfully with bouncy steps and arm waves. No camera movement. Cut to Shot 2 (4-7s): medium shot from a slightly low angle, faces clearly visible and sharp, smooth sideways track along the row as each dancer does a quick signature move (spin, hair flip or shoulder shimmy) and the coats swing. Cut to Shot 3 (7-10s): medium-wide shot, slow push-in while all four bounce together, then strike a happy group pose and hold it for the last half second. Style: loose, energetic fashion-dance video with hair tosses, arms up, shoulder shimmies, playful poses, laughter and buoyant group energy, similar to a casual studio dance montage but outdoors. Keep each dancer's identity and outfit consistent in every shot. Audio: upbeat, catchy instrumental background music throughout, pop-funk groove with claps and bright guitar, plus light street ambience. No speech, dialogue, singing or lip syncing. No captions, subtitles, titles, labels or watermarks.

One single continuous take of exactly 30 seconds, vertical 9:16, ZERO cuts, no edits, no fades. The video opens on the supplied start frame (Source A, the first frame): a low wide view from the sidewalk, the closed door of the white wooden house in the upper half, and the three kids seen from behind on the path walking toward the porch, ghost left, skeleton centre carrying the orange bucket, witch right. Keep the strong, visible 1980s analog 35mm film grain of the start frame throughout. Keep all key action, faces, hands, candy and bucket inside the central safe area of the vertical frame with generous margins. Match all supplied references exactly. Source B: Ghost kid costume and identity (white bedsheet with eye holes, white sneakers; face never visible). Source C: Skeleton kid costume and identity (bone-print jumpsuit, hood up, printed gloves, skull makeup). Source D: Witch kid costume and identity (bent pointed hat, tulle dress, striped tights, lace-up boots, braids). Source E: orange plastic jack-o'-lantern candy bucket. Source F: hero candy, orange metallic foil body, purple foil twisted ends, no text. Source G: 1985 suburban house, porch, steps, door layout. Source H: the woman who opens the door, face, permed 80s hair, oversized orange Halloween-print wool sweater, jeans. Subject: exactly three children aged 8-10 and exactly one adult woman, no extra people. Ghost kid on the left, skeleton kid in the centre (black hood up, white skull makeup, carries the orange jack-o'-lantern bucket by its black handle), witch kid on the right (slightly bent black pointed hat, black tulle dress, purple-and-black striped tights, black lace-up boots, two long brown braids, no cape, empty hands). The woman is about 35-40, warm friendly face, voluminous permed 1980s hair, bright 80s eyeshadow and blush, large earrings, high-waisted light-wash jeans, oversized chunky orange wool sweater with a knitted bats, ghosts and pumpkins print. The bowl holds individually wrapped, invented, unbranded, text-free candies. The single hero candy is a hard candy about 4 cm long, orange metallic foil body with purple twisted ends, no text. Timeline, in this exact order: 0:00-0:04 APPROACH: the three kids walk up the path toward the closed door, seen from behind. One smooth steadicam-style forward dolly follows them from knee height and ends in a medium-wide view of the porch with the kids at the foot of the steps. Deep focus. Footsteps and dry leaves, soft evening ambience. 0:04-0:06 DOORBELL: the skeleton kid presses the lit doorbell. A clear, unmistakable classic two-tone "ding-dong" rings from inside the house. The camera settles into a medium-wide framing, kids at the bottom, door in the upper half. 0:06-0:09.5 DOOR OPENS AND LINE: the door opens with a soft creak and stays open for the rest of the video. The woman appears in the doorway, beams warmly and says live on camera, with natural lip movement, in a cheerful natural American English voice, exactly: "Oh, hello kids, do you want some candy?" Make the English dialogue audible exactly as written. Slight push toward the doorway; focus eases to her face and then returns to deep focus. 0:09.5-0:11 KIDS REACT: all three kids jump with joy and shout together in unison, in happy children's voices, a long excited "Yeeeesss!" The ghost bounces, the witch cheers, the skeleton kid's painted face lights up with delight. Medium-wide, no cut. 0:11-0:13.5 HANDFUL: the woman picks up the white ceramic bowl from the small table just inside the doorway and takes a big handful of wrapped candy in her fist. Medium framing. 0:13.5-0:15 BUCKET: the skeleton kid steps forward and holds the open orange pumpkin bucket directly beneath her closed fist, level with her hands. The camera moves slightly closer to her hand and the bucket. 0:15-0:16 FIST OPENS: she opens her fist over the bucket and the candies begin to fall at real speed. 0:16-0:22 BULLET TIME: within the same unbroken take, time slows to extreme super-slow motion, about one-thirtieth normal speed. The candies hang almost frozen in the air, wrappers glint, her open hand and the kids' gloves are nearly still. Perform one slow, smooth full 360-degree orbit at candy height around one single falling hero candy. It is the only razor-sharp element, in shallow focus, foil highlights glinting as it turns. In the softly out-of-focus background, the skeleton kid's face, white skull makeup and black hood visible, gazes up at the falling candies with wonder and excitement, eyes wide and mouth open in a delighted smile; as the camera circles, the blurred face stays in the background. A low time-slow whoosh; all other sound is muffled. 0:22-0:25 REAL TIME AND OVERHEAD: time snaps back to normal speed and sound returns. The camera rises smoothly in one crane-up move to a brief top-down, straight-down view directly above the open pumpkin bucket. The remaining candies fall into it at real speed, bouncing and rattling into the plastic bucket with real gravity, no cartoon bounce. Keep the bucket centred. 0:25-0:27 DESCENT: the camera descends smoothly back to a front medium-wide framing of the porch with no cut: the three kids at the foot of the porch steps facing the door, the skeleton holding the now-full bucket, the woman in the open doorway beaming at them. 0:27-0:30 KIDS TURN AND RUN AWAY: the three kids, delighted, turn around together, show happy big smiles, and run away down the porch steps and back along the path toward the sidewalk, the ghost's sheet flapping, the witch's braids bouncing, the skeleton holding the full bucket, candy rattling inside it. Quick excited footsteps on the wooden steps and dry leaves, costume rustle. They leave the frame while the woman stays standing calmly in the open doorway at the threshold, smiling warmly after them, without waving and without moving out of the doorway. Medium-wide camera held steady, no cut. Environment: match Source G and the established setting: 1985 American suburban house with white wooden siding, front porch with three steps, brass knocker, glowing carved jack-o'-lanterns on the steps, dry autumn leaves, cotton cobwebs on the railing, short path from the sidewalk to the porch. A white ceramic bowl full of wrapped candy sits on a small table just inside the doorway until the woman picks it up. Warm 2700K porch-light key, cool rear fill, subtle orange pumpkin flicker at knee height. Camera: only smooth, stabilized, motivated moves: the opening forward dolly, slight push toward the doorway, slight move closer to the hand and bucket, one full 360-degree orbit around the single hero candy, one smooth crane-up to the brief overhead view, one smooth descent, then a steady held medium-wide. No handheld shake, no zoom. Deep focus except for the bullet-time shallow-focus orbit. Style: friendly 1985 American sitcom Halloween scene, warm early-evening dusk, soft diffusion and halation, saturated faded warm telecine colour, 35mm network-broadcast film look with clearly visible fine film grain, slightly soft resolution, welcoming and evenly lit rather than noir. No brands, logos or licensed characters. Audio: sound effects and natural ambience only: footsteps, dry leaves, soft evening ambience, the clear two-tone doorbell, soft door creak, the woman's spoken English line, the children's unified "Yeeeesss!", candy rattling in plastic, the muffled bullet-time whoosh, running footsteps on the steps at the end. ABSOLUTELY NO MUSIC at any point: no tune, no score, no melody at the start, in the middle or at the end. No voice-over, narration, singing or laugh track. Constraints: no extra people. No cape on the witch. The ghost's face is never visible. The witch's hands stay empty. The skeleton carries the bucket. The door remains open from 0:06 to the end. The hero candy stays one single orange foil candy with purple twisted ends and no text. No legible text, captions, titles or watermarks. Start exactly on the supplied start frame (Source A).

Tiempo (Seg) Plano y Encuadre Acción y Actitud (100% Animales de Plastilina) Audio Desglosado 00:00 - 00:04 Plano Medio (Exterior / Escaleras) El Perro con gafas camina por la calle nocturna bajo la lluvia, ve el letrero de neón "MUV", empuja la puerta y baja las escaleras oscuras al sótano. Su abrigo brilla con purpurina y corbata fucsia de leopardo. JPG 00:00 - 00:03: Pasos + Jazz tenue/sordo de fondo. 00:04: Golpe seco de puerta → Explosión de sonido nítido. 00:04 - 00:08 Primer Plano (Ni bien llega) El Tigre (Saxofón - Sexy): Lo primero que ve el Perro al pisar el sótano. Viste chaleco de leopardo, sostiene una nota con pasión y le guiña un ojo a la cámara con una sonrisa magnética. JPG 00:04 - 00:08: Solo triunfal, brillante y coqueto de Saxofón. PNG 00:08 - 00:12 Plano General (La Banda y Público) Escenario del sótano kitsch. Se revela la banda completa y a los espectadores (todos animales: ratones, cerdos, conejos con trajes elegante/kitsch, sin humanos) escuchando el Swing. PNG + 1 00:08 - 00:12: Entra toda la base rítmica de Gypsy Swing (Batería, Violín, Contrabajo). PNG 00:12 - 00:16 Plano Detalle (Batería) El Mono (Batería - Pícaro): Sonríe de lado con su corbatín torcido, tocando a toda velocidad con sus brazos de plastilina sobre los platos y la caja. JPG + 1 00:12 - 00:16: Destaque de Batería (repiques rápidos de escobillas y platillos). PNG 00:16 - 00:20 Plano Medio (Violín) El Oso (Violín - Melancólico): Rostro triste y expresivo, tocando su diminuto violín con profunda nostalgia. JPG + 1 00:16 - 00:20: Solo dramático y melódico de Violín. PNG 00:20 - 00:25 Primer Plano (Contrabajo) El Gato (Contrabajo - Cool): Párpados caídos, actitud relajada, con anillos gigantes de gemas en sus patas mientras pellizca las cuerdas gruesas. JPG + 2 00:20 - 00:25: Solo grave de Contrabajo (walking bass marcando cada segundo). PNG 00:25 - 00:30 Plano Medio (Cierre) El Perro se acomoda en la barra junto a un barman bulldog y una lámpara de flamenco rosa, levanta su vaso de plastilina en aprobación y asiente al ritmo. Fundido a negro. JPG 00:25 - 00:28: Subida musical en conjunto (crescendo). 00:29 - 00:30: Acorde Tutti final con remate seco de plato y silencio. PNG

Car-to-car tracking shot running parallel to a cherry-red vintage convertible with a tan interior, top down, crossing the Golden Gate Bridge left to right, the bridge cables sweeping past behind it, sailboats on the bay far below, warm golden late-afternoon light. Standard normal lens character with a 47-degree diagonal field of view, held constant with no field-of-view drift. Dex drives, relaxed, one wrist resting on the wheel. Pete sits in the front passenger seat and waves warmly straight at the camera. In the back seat, Frankie kneels up on the seat waving with her small orange-haired toy figure in her fist, pigtails whipping in the wind, laughing; Marv points finger-guns at the camera. All four look directly at the camera and smile with open, direct opening-credits energy. Exactly these four people are in the car, no extras. Wind moves hair and collars naturally; the convertible rides with real suspension sway.

Candid lifestyle product photograph. An older silver-haired man with a weathered, characterful face wearing large tinted sunglasses, drinking from a straw inserted into a yellow MIT() soda can held in his hand. He is wearing a dusty rose pink MIT() hoodie with the cream "MIT()" logo clearly visible on the chest. Medium close-up shot, slightly low angle. Sunny coastal outdoor setting — blue sky and water in the background, warm golden natural light. Relaxed, confident, effortlessly cool expression. The yellow MIT() can with red "MIT()" lettering and "SPARKLING ENERGY ZERO SUGAR" text is clearly visible. Candid editorial lifestyle photography style. No extra text or graphics. Do not reproduce incidental presentation headers, view names, palette labels, captions, annotations, borders, grids, or layout from the reference images. Preserve writing from a reference only when this prompt explicitly requires that exact in-world text or graphic.

Cinematic film still, 1990s sitcom living room, a couple in their early thirties sitting side by side on a cream fabric sofa, sharing a big laugh together. Left: a woman with shoulder-length brown hair and straight bangs, lilac cardigan over a white T-shirt, head tipped back with joyful laughter, one hand near her cheek. Right: a man with flat combed brown hair, beige crew-neck sweater, laughing broadly with his mouth open, one hand patting his own knee in amusement. Lavender wall and oak bookshelf behind them, open kitchen visible in the background. Three-point multi-camera sitcom lighting, soft frontal key from high, generous fill, 3200K tungsten warmth, even exposure, no hard shadows. Palette 60% warm oak and cream, 30% lavender purple, 10% red and pink accents. Shot on 35mm broadcast film, fine grain, slightly soft, saturated telecine color, authentic 1994 network sitcom look. Do not reproduce incidental presentation headers, view names, palette labels, captions, annotations, borders, grids, or layout from the reference images. Preserve writing from a reference only when this prompt explicitly requires that exact in-world text or graphic.

One continuous 12-second, ultra-photorealistic live-action shot, styled as a high-budget fashion campaign / cinematic music video with real actors, practical props, real high-speed photography. Environment: a San Francisco street inspired by the reference photo — steep sloped street with cable-car tracks, tall mixed-era skyscrapers, warm late-afternoon light — new camera angle, open pedestrian space for the walk and collision; pedestrians stay in pedestrian areas, vehicles in vehicle areas, realistic scale. CRITICAL PHONE RULE: there is exactly ONE phone in this whole video: a compact early-2000s silver-gray OPEN CLAMSHELL FLIP PHONE — two hinged rectangular halves, small illuminated display and speaker on top, physical numeric keypad and microphone below, visible central hinge, already open. This exact flip phone is in the protagonist's RIGHT HAND from the very first frame, held to the right ear during a call, and it is this SAME phone that later releases from the hand. No modern smartphone or touchscreen phone ever appears, in this hand or anywhere, at any camera distance. Wide, medium, and macro shots all show this one identical device. CENTRAL CONCEPT: a dramatic bullet-time accident. At the collision, objects get one short realistic release impulse each, then time drops to about 2-3% speed. Released objects become nearly stationary fixed points in 3D world space while only the CAMERA moves through the frozen moment to inspect them — objects never fly toward the lens or self-enlarge; only reduced camera distance creates each close-up. Real-world physics, causality and spatial continuity are critical. EXACT COUNTS: one protagonist, one secondary passer-by, one Labrador on one leash, one MITO Clubmaster sunglasses (hero object), one coffee cup with plastic lid, one silver-gray open flip phone as above. No duplicate or different props anywhere. HAND OWNERSHIP before collision: RIGHT HAND holds the same open flip phone (small MITO logo) to the right ear, mid call, natural subtle lip/jaw movement. LEFT HAND holds the paper coffee cup (small MITO logo), lid attached, coffee inside. The MITO Clubmaster sunglasses are worn on the face. 0.0-2.0s REVEAL: extreme low close-up on the protagonist's red high-top Converse walking; one continuous smooth crane-up plus gentle backward tracking reveals in order: red Converse -> relaxed blue jeans -> white "I ❤️ SF" t-shirt -> torso -> MITO Clubmaster sunglasses on the face -> full protagonist, still on the same flip phone, coffee in left hand. 2.0-3.7s WALK: smooth frontal tracking continues; exactly one secondary person walking one Labrador on a leash approaches from the opposite direction (dog slightly ahead). They stay supporting elements. Identity, wardrobe, sunglasses, the same flip phone, and coffee stay fully stable — phone never changes shape or type. 3.7-4.4s COLLISION: the secondary person's elbow accidentally clips the protagonist's arm/torso — small, believable contact. The protagonist recoils backward. Three release events from this same collision: (1) the sunglasses get one short outward impulse and separate from the face, traveling only a short distance, never toward the camera; (2) the coffee cup leaves the left hand with a short outward/downward rotational impulse, tilts and rotates, and rotation plus liquid inertia pops one edge of the lid loose, opening a real gap through which coffee only then begins escaping, slow and restrained; (3) the SAME flip phone leaves the right hand with a short outward/upward impulse, keeping its exact design and open angle — the identical device held since frame one, never a different phone. The instant each object leaves, that hand is completely empty — nothing else ever appears in the right hand before or after that release. The protagonist's face and eyes are already complete under the sunglasses and are revealed normally once they separate: stable pupils, natural eyelids, correct gaze, symmetrical brow/nose/cheekbones. Hair keeps its shape with only subtle natural inertia. 4.4-5.2s FROZEN WORLD: time drops to ~2-3% speed. First hold a clear medium-wide view of the whole accident before any macro inspection — protagonist beginning to fall backward, both hands empty, sunglasses suspended a short distance from the face, coffee cup suspended lower toward the left-hand side with its partly detached lid and a small escaping stream, and the same open flip phone suspended on the opposite right-hand side. These become fixed world-space anchors; objects drift only microscopically and never approach the camera or grow in frame — the camera alone moves from here on. 5.2-6.8s HERO OBJECT: camera dollies toward the nearly stationary sunglasses, slows, holds, then a restrained 20-30 degree orbital parallax around them. Reveal the glossy black acetate, gold-tone hinge rivets, small "MITO" wordmark beside the rivet, real material thickness, edges, reflections, shallow depth of field, falling protagonist softly visible behind. Must read as a real product shot with a macro cinema lens, never CGI. Complete before moving on. 6.8-8.3s COFFEE: camera travels diagonally to the same cup, holds a macro close-up: same paper cup, lid partly detached along one edge, a thin irregular stream and a few droplets escaping through the real gap, natural surface tension and reflections — no explosion, no CGI splash. Small lateral parallax move, then leave. 8.3-9.8s PHONE: camera travels to the same suspended flip phone — the identical device since frame one, never a smartphone — first holding its complete silhouette with hinge and keypad to confirm it's the same device, then moving closer to the display for an extreme macro close-up where the illuminated screen reads "BOSS" with a small phone number beneath, real surface wear and reflections. The right hand stays empty in the background. The phone never becomes a smartphone or a different phone. 9.8-10.4s RECONSTRUCT: a fast, smooth backward dolly reveals the complete frozen accident again — protagonist still falling, both hands empty, the sunglasses, the cup with its droplets, and the same flip phone all still at their established positions, plus the secondary person and Labrador — nothing teleported, regenerated, or changed type. 10.4-12.0s PAYOFF: time snaps back to normal speed; gravity and momentum resume together. The protagonist completes the fall. The sunglasses continue from their position and land under their own momentum. The coffee cup continues, hits the ground, remaining coffee spills irregularly from the real opening/impact point. The same flip phone continues, strikes the ground, makes one small believable bounce. The three objects land at different plausible spots, not converging beside the protagonist. Camera tilts/drops to ground level for a low-angle finish connecting the fallen protagonist and scattered objects on the same street. Hard cut to end. Every visible change needs a physical cause; nothing transforms just because the camera gets closer; no prop is duplicated or regenerated; the phone is one silver-gray flip phone from first frame to last, never a smartphone.

Cinematic dark drama portrait shot, 35mm film aesthetic with heavy grain, deep cool color grading, desaturated tones, and soft overcast daylight under gentle rainfall. Aspect ratio 9:16 vertical, 10 seconds duration. [00:00 - 00:04] Intimate medium close-up shot of a pale young woman with wet hair, wearing a thick grey wool coat in the rain. A swarm of tiny, adorable fluffy kittens (calico, ginger, black, and tabby) are snuggled closely around her neck, chest, and shoulders, clinging to her coat. Her face begins with a tearful, emotional expression as rain trickles down her cheeks. [00:04 - 00:07] Slowly, a gentle, tender smile breaks across her wet face as she feels the warmth of the kittens, her eyes softening as water droplets glisten on her skin. [00:07 - 00:10] The tiny kittens squirm, meow softly, and nuzzle against her chin while she takes a deep breath, her subtle smile fading into a quiet, vulnerable, and reflective gaze into the distance.

A single unbroken 30-second continuous tracking shot moves fluidly through a lovingly cluttered warm suburban cottage garden at golden hour, weaving between locations and characters with smooth camera glides and pushes — no hard cuts, no jump cuts, realistic handheld-documentary camera language throughout. Maintain exact, unwavering visual consistency of both the gnome and the mailman across every beat: identical proportions, colors, materials, and clothing from start to finish, with no drift in their design. 0:00-0:03 — The camera glides forward in a slow push over the garden: mismatched vintage brass lanterns, chipped porcelain teacups used as planters, a striped red-and-cream pinwheel spinning lazily, an old rusty bicycle wheel with wildflowers growing through its spokes. Ambient sound only: wind chimes, birds, a soft breeze. 0:03-0:05 — The camera drifts smoothly across a row of ordinary decorative garden gnomes lined up along the flowerbed — each sculpted in a similar terracotta style with small knitted hats, completely motionless — and holds a beat too long on one of them: the protagonist gnome, with faded warm terracotta skin, oversized soulful brown eyes, a floppy hand-knitted crimson-red pointed hat, and a knitted moss-green tunic. His eyes flick subtly sideways and the brim of his hat gives a barely perceptible twitch. Score drops to near silence. 0:05-0:08 — The camera pushes into an extreme close-up on his eyes narrowing with suspicious determination, same exact face and hat design. Score kicks in low and ominous. Gravelly whispered voiceover: "Day 47. He's back." 0:08-0:11 — The camera pulls back and glides toward the garden path in slow motion, revealing the friendly postal worker — same navy uniform, canvas mail satchel, and face as established — walking calmly up the path, while the protagonist gnome peers out from behind a weathered terracotta flowerpot with a tiny brass magnifying glass, his design unchanged. Gravelly whispered voiceover: "Same time. Same route. That's exactly what they WANT you to think." 0:11-0:15 — The camera moves fluidly to the gnome's miniature conspiracy evidence board — thin wooden stakes strung with red twine, a photo of a squirrel, a real acorn, a small sprinkler timer diagram, dried leaves, tiny handwritten notes — as he frantically pins a fresh leaf to it, his appearance still identical to every earlier beat. Gravelly whispered voiceover: "Sprinklers went off four minutes early today. Coincidence? I don't think so." 0:15-0:18 — The camera glides back toward the cream front door as the postal worker, same uniform and satchel, drops a stack of mail through the brass mail slot, completely normal and unhurried, while the gnome gasps and dives out of sight behind a nearby flowerpot. Gravelly whispered voiceover: "He knows that I know." 0:18-0:22 — The camera pushes into a close-up on the gnome, same consistent design throughout, standing beside the pile of delivered mail. A plain unbranded colorful flyer flutters loose from the pile. His eyes go huge with sincere existential alarm, freezing in a startled pose. The frame freezes, and a bold vintage rubber-stamp graphic slams down across the image in distressed red stamped ink, reading "CASE UNSOLVED" — a single intentional in-world stamp graphic, with a satisfying thud sound effect on impact. Throughout: rich saturated retro postcard color grade, warm golden-hour light, analog 35mm film texture, dry mockumentary comedy tone, continuous unbroken camera movement connecting every beat, consistent character design from first frame to last. No readable text anywhere except the intentional "CASE UNSOLVED" stamp in the final freeze frame.

Camera Angle & Perspective: Strict first-person perspective (POV shot). A single, continuous fluid camera movement with zero cuts, simulating a human eye view. Hand motion must maintain realistic human anatomy: exactly five fingers, correct joint proportions, natural wrist rotation, and no floating or warping artifacts. Setting & Environment: Inside a dimly lit, high-end tech conference auditorium filled with a seated audience. In the background, a massive glowing presentation screen displays a male speaker in a black sweater against a clean white backdrop. Action Breakdown (Second-by-Second Flow): The Toss (0:00 - 0:02): From the POV angle, look toward the large illuminated stage screen. On screen, the male presenter smiles, raises a sleek light-gray folded smartphone, and throws it forward toward the camera in a smooth motion. Mid-Air Flight & Unfolding (0:02 - 0:04): The camera tilts slightly up as the physical foldable phone flies out from the screen area, spinning through the dark air of the auditorium. Mid-flight, the device smoothly unfolds 180 degrees into a wide inner screen displaying a clear iOS-style home interface with vibrant app icons, a photo widget, and a clock. The Catch & Audience Shock (0:04 - 0:07): The camera moves downward to track the landing. A single, perfectly rendered right human hand reaches up into the lower frame. The open smartphone drops seamlessly and gently into the palm. Adjacent audience members in the surrounding seats turn their heads toward the hand in genuine shock and jaw-dropping amazement, laughing and reacting naturally. Inspection & Folding (0:07 - 0:10): The hand holds the unfolded phone steady for a fraction of a second, showing the sharp screen display. The hand then closes the phone shut with a firm, clean snap. The camera focuses on the folded exterior: a matte off-white rear glass casing with a dual-camera module on top and an Apple logo centered. Lighting, Texture & Rendering Style: Photorealistic 8K render, dramatic cinematic key lighting from the main stage, natural motion blur during the aerial throw, shallow depth of field focusing sharply on the handheld phone while keeping the background audience slightly soft. Clean geometric symmetry on the smartphone body, zero texture flickering or digital distortion. High frame rate (60fp Source B

Physical tabletop stop-motion animation shot at 12 fps, frame-by-frame photographed. Motion is visibly stepped and punchy, pieces snap between discrete positions. No interpolated smoothness, no CGI animation. Flat solid vivid-blue seamless background, crisp hard studio lighting with defined shadows. No people, no hands. Multi-shot advertising edit: each shot has a LOCKED camera, shots change only through hard cuts. The reference image defines the exact waffle stack's texture, syrup, butter, powdered sugar, and the MITO branding on the paper flag — strictly no other logos or third-party branding anywhere. Preserve that identity exactly, do not redesign it. SHOT 1 (0-2s) locked overhead wide: the waffle stack completely disassembled and spread out — three separate golden Belgian waffle squares, a pat of butter, a dusting of powdered sugar, a pool of syrup, and the kraft-paper flag printed with the "MITO" logo — each piece jittering into a new position every few frames on the solid vivid-blue background under studio lights. SHOT 2 (2-4s) hard cut to low frontal: the bottom waffle square slides into center and settles, the second waffle square drops and stacks on top in a stepped snap. SHOT 3 (4-6s) hard cut to macro three-quarter: the third waffle square snaps onto the stack, the pat of butter drops onto the top, and the powdered sugar dusts down in sharp increments. SHOT 4 (6-8s) hard cut to locked overhead: the syrup drizzles down over the stack in one sharp stepped stream, then the kraft-paper "MITO" flag spikes down and plants into the top waffle. SHOT 5 (8-10s) hard cut to frontal hero framing: the finished waffle stack rotates through four discrete stepped positions on the vivid-blue background, landing in the exact editorial hero pose.

Physical tabletop stop-motion animation shot at 12 fps, frame-by-frame photographed. Motion is visibly stepped and punchy, pieces snap between discrete positions. No interpolated smoothness, no CGI animation. Flat solid mustard-yellow seamless background, crisp hard studio lighting with defined shadows. No people, no hands. Multi-shot advertising edit: each shot has a LOCKED camera, shots change only through hard cuts. The reference image defines the exact hot dog's bun, sausage, mustard, relish, onion, and the MITO branding on the kraft-paper wrap — strictly no other logos or third-party branding anywhere. Preserve that identity exactly, do not redesign it. SHOT 1 (0-2s) locked overhead wide: the hot dog completely disassembled and spread out — the toasted bun's top and bottom halves, the grilled all-beef sausage, a squeeze line of mustard, relish, diced onion, and the kraft-paper wrap printed with the "MITO" logo — each piece jittering into a new position every few frames on the solid mustard-yellow background under studio lights. SHOT 2 (2-4s) hard cut to low frontal: the bun halves slide together and settle open, the grilled sausage drops into the split and snaps into place. SHOT 3 (4-6s) hard cut to macro three-quarter: the mustard zigzags across the sausage in a stepped squiggle, then the relish and diced onion snap into place one after another. SHOT 4 (6-8s) hard cut to locked overhead: the kraft-paper wrap folds up and snaps around the base of the hot dog in sharp increments, the "MITO" mark now visible on the front. SHOT 5 (8-10s) hard cut to frontal hero framing: the finished hot dog rotates through four discrete stepped positions on the mustard-yellow background, landing in the exact editorial hero pose.

Physical tabletop stop-motion animation shot at 12 fps, frame-by-frame photographed. Motion is visibly stepped and punchy, pieces snap between discrete positions. No interpolated smoothness, no CGI animation. Flat solid dark green seamless background, crisp hard studio lighting with defined shadows. No people, no hands. Multi-shot advertising edit: each shot has a LOCKED camera, shots change only through hard cuts. The reference image defines the exact burger's bun, patty, cheese, lettuce, tomato, sesame seeds, and the MITO branding on the cheese slice and paper flag topper — strictly no other logos or third-party branding anywhere. Preserve that identity exactly, do not redesign it. SHOT 1 (0-2s) locked overhead wide: the burger completely disassembled and spread out — sesame bun top, bun bottom, beef patty, cheese slice with the "MITO" mark, lettuce leaf, two tomato slices, and the kraft-paper MITO flag topper — each piece jittering into a new position every few frames on the solid dark green background under studio lights. SHOT 2 (2-4s) hard cut to low frontal: the bun bottom slides into center and settles, the patty drops and snaps down onto it in a stepped motion. SHOT 3 (4-6s) hard cut to macro three-quarter: the cheese slice with the "MITO" mark leaps in and drapes over the patty in stepped snaps, then the lettuce leaf and tomato slices snap into place one after another, one per frame. SHOT 4 (6-8s) hard cut to locked overhead: the sesame bun top descends and snaps onto the stack; the kraft-paper MITO flag topper spikes down and plants into the bun in one sharp stepped motion. SHOT 5 (8-10s) hard cut to frontal hero framing: the finished MITO burger rotates through four discrete stepped positions on the dark green background, showing only the MITO branding on the cheese and flag topper, landing in the exact editorial hero pose.

Continuous single-take, ultra-photorealistic live-action fashion-campaign short, feels like a real high-budget commercial shot on premium cinema lenses with practical props and real high-speed photography — never CGI product animation. Cast and props (no duplicates, ever): one protagonist matching the identity reference, wearing the MITO staff ID card on its taut lanyard around his neck (the hero object, bearing the "MIT()" mark), one closed silver laptop with one tablet and one stylus stacked cleanly on top, and one black coffee mug printed "we are searching for talent," filled with coffee. Before the trip, his left arm cradles the laptop stack and his right hand holds the coffee mug; the ID card stays on its lanyard throughout and is never held. 0.0–2.0s: Extreme low close-up on the protagonist's footwear walking through a bright domestic hallway — white walls, warm wood floor, hard direct sunlight through waist-height residential windows on camera-left, warm light rectangles across wall and floor, dust motes visible in the sun shafts. One continuous crane-up + gentle backward track reveals footwear → legs → clothing → torso → ID card and lanyard → face, landing on a near full-body composition. He's clearly recognizable, ID card swinging gently against his chest, laptop stack under his left arm, coffee mug in his right hand. 2.0–3.7s: Smooth frontal tracking continues as he walks on; the hallway is background texture only — straight architectural lines stay perfectly rectilinear, classic wide-angle (~84°), no fisheye curve. He approaches a floor edge/rug transition ahead. 3.7–4.4s: His shoe catches the floor edge and he trips, launching forward with a startled recoil. The laptop-tablet-stylus stack leaves his left arm with one short outward impulse, the tablet and stylus separating just slightly from the laptop but never duplicating; the coffee mug leaves his right hand with an outward/downward rotational impulse, and only because of that rotation and the coffee's inertia does a single continuous stream begin to arc free of the mug — no explosion, no decorative splash. The ID card swings outward on its taut lanyard, still around his neck. Both hands are instantly and fully empty. His face and hair stay anatomically stable and continuous — nothing regenerates when the props separate. 4.4–5.2s: Time drops to roughly 2–3% speed. Hold a medium-wide view of the whole frozen stumble first: he's mid-fall, both hands empty, the ID card suspended out from his chest on its taut lanyard, the laptop-tablet-stylus stack suspended to one side with the stylus and tablet just barely separated from the laptop, the coffee mug tilted lower on the other side with one frozen ribbon of coffee arcing free. These become fixed world-space anchor points — objects barely drift; the camera does all the moving from here on. 5.2–6.8s: The camera dollies toward the almost-stationary ID card (it grows in frame only because the camera gets closer, never by moving toward it), holds, then performs a restrained 20–30° orbital parallax around it, revealing the card face, lanyard, and the "MIT()" mark in shallow macro depth of field with him still softly visible behind. 6.8–8.3s: The camera leaves the card in place and travels to the suspended coffee mug. Macro hold on the mug and its printed "we are searching for talent" text: the single frozen coffee stream and a few individually sized droplets hang mid-air. A small lateral parallax move studies the ceramic surface, the lettering, and the coffee before the camera moves on. 8.3–9.8s: The camera leaves the mug in place and travels to the suspended laptop-tablet-stylus stack, holding briefly on the full silhouette to confirm it's the same clean, undamaged silver laptop before moving to an extreme macro on the tablet screen and the stylus spinning slowly free beside it. Hold on real housing texture, screen reflections, and the stylus's fine detail. His hands stay empty in the soft background. 9.8–10.4s: A fast, smooth dolly back reveals the complete frozen tableau again — he's still falling, both hands empty, ID card, mug with its coffee stream, and laptop stack all exactly where they were, nothing teleported or regenerated. 10.4–12.0s: Time snaps back to full speed. Gravity and existing momentum resume for every object from its own suspended position, not from his arms: he finishes falling forward; the ID card swings down and comes to rest against his chest as he lands; the laptop lands separately with the tablet and stylus sliding off but staying distinct, single items; the coffee mug hits the floor and the remaining coffee spills irregularly across the wood. The laptop stack and mug end up in different, physically plausible spots — they do not converge back next to him. The camera drops to a low angle following the impact, ending on a cinematic low composition connecting him and the scattered props. Hard cut. Camera and physics rules: bullet time means the physical world nearly freezes while only the camera moves through it — never move an object toward the lens to fake a close-up, never enlarge or teleport an object. Every visible change has a physical cause (the stumble, released momentum, resumed gravity) — nothing transforms just because the camera gets closer. The macro inspections revisit the exact same physical props released earlier at different camera distances, never redesigned or duplicated versions of them. No extra readable text beyond the "MIT()" mark on the ID card and the "we are searching for talent" print on the mug.

Style & Cinematography: Cinematic, high-concept commercial directed in modern Apple/tech aesthetic, shot on 35mm lens, photorealistic 8k, vibrant lighting, shallow depth of field, natural motion blur. Main Character: Jaume, a 28-year-old Mediterranean guy with short brown hair, wearing a green jacket over a white collared shirt with a tie. Desglose por Tomas (Seq. Shot-by-Shot) Shot 1 (Café Intro): Medium close-up of Jaume sitting inside a sunlit, modern glass café. He holds up a sleek white dual-screen folding smartphone, showing it off enthusiastically to the camera with a joyful smile. Shot 2 (Surreal Morphing): Quick whip-pan camera move toward the coffee counter. The barista transforms into a standing, human-sized golden fish wearing a formal suit and tie. The fish loses balance and tumbles onto the tiled floor. Shot 3 (Street Chase & CGI Elements): Dynamic POV camera rushing out of the café into a busy city street. Panicked crowd running. A giant green praying mantis appears in first-person perspective, followed by glowing green wireframe digital sea-horses galloping through green matrix-structured buildings. Shot 4 (Inception Gravity Shift): Fast tracking shot down the avenue behind running pedestrians. The cityscape inverts, bending the horizon upside down in an Inception-style gravity warp under a dramatic sky. Shot 5 (Wake Up): Sudden cut to a close-up of Jaume gasping awake on a grey sofa in a brightly lit studio room, wearing a grey cardigan over a white shirt. Shot 6 (Cat Scene): Camera pans to a wooden desk. A grey Scottish Fold cat wearing a backward red cap, blue glasses, and a red plaid shirt is sitting while operating the dual-screen folding phone showing a goldfish animation. Jaume approaches from behind, smiling warmly and petting the cat. Shot 7 (Flatlay outro): Overhead top-down shot of a wooden desk setup featuring the dual-screen phone open next to an Apple logo, wireless earbuds, smart watch, and technical drawings. Warm morning sunlight casting soft shadows. Source B

Physical tabletop stop-motion animation shot at 12 fps, frame-by-frame photographed. Motion is visibly stepped and punchy, pieces snap between discrete positions. No interpolated smoothness, no CGI animation. Flat solid dark green seamless background, crisp hard studio lighting with defined shadows. No people, no hands. Multi-shot advertising edit: each shot has a LOCKED camera, shots change only through hard cuts. The reference images define the exact pink and red suede/canvas sneaker design and materials. All branding must exclusively feature "MITO" on the tongue, side stripes, and heel tab—strictly no Adidas logos or third-party branding anywhere on the shoe or heel. Preserve that identity exactly, do not redesign it. SHOT 1 (0-2s) locked overhead wide: the pink sneaker completely disassembled and spread out—rubber sole, pink upper panels, tongue with MITO logo, heel tab with MITO logo, insole, metal eyelets, two loose red flat laces—each piece jittering into a new position every few frames on the solid dark green background under studio lights. SHOT 2 (2-4s) hard cut to low frontal: the sole slides into center and settles, pink upper panels fold up and construct themselves around it in stepped snaps. SHOT 3 (4-6s) hard cut to macro three-quarter: the tongue with the "MITO" logo leaps in and drops into the throat, eyelets snap into the panels one after another, one per frame. SHOT 4 (6-8s) hard cut to locked overhead: with the pink sneaker now completely assembled, the two loose red laces snake across the solid dark green background in sharp stepped waves toward the built shoe. SHOT 5 (8-10s) hard cut to extreme macro on the eyelets: the red laces thread themselves through every eyelet one at a time in visible increments, cross over, then whip up and tie into a perfect symmetrical bow, loops popping into shape. SHOT 6 (10-12s) hard cut to frontal hero framing: the finished pink and red MITO sneaker rotates through four discrete stepped positions on the dark green background, showing only MITO branding on the tongue, sides, and back, landing in the exact editorial hero pose