Референсные кадры в ИИ-видео: как начальное изображение, промпт и модель формируют план

Use the attached image as the exact starting frame reference.

Create a 4-second ultra-realistic cinematic video in 9:16 vertical format showing the same six women running fast along Burlingame Avenue in a competitive street race.

Scene and continuity
Keep the same street, the same perspective style, and the same environment from the reference image:
- Burlingame Avenue
- same storefront area and architecture
- same trees, banners, lamps, sidewalks, parked cars, café area, pedestrians
- same daylight and overall look
The video should begin from the attached frame and continue naturally from it.

Camera movement
The camera should move alongside the runners, traveling in the same direction as their run, as a side-tracking shot. The feeling should be like a camera operator or tracking rig moving parallel to them on the street.
Important camera behavior:
- smooth lateral tracking
- no shaking
- no sudden jumps
- no zoom in/out
- no spinning
- no unnatural lens distortion
- keep the runners framed from the side / slight three-quarter side angle
- the camera speed should match their movement so they remain the main focus while the background shifts naturally

Action
The six women are running with real competitive intensity, not casually jogging.
Keep the same six athletes and outfits:
1. light blue patterned STRONGER set
2. dark blue STRONGER oversized shirt and shorts
3. black Nike sports bra and leggings
4. white Adidas top with green pants
5. pink HOKA sports bra and shorts
6. white / gray New Balance top, jacket, and black shorts
They should run in a natural staggered formation, not in a straight line:
- some slightly ahead
- some slightly behind
- irregular spacing
- subtle overtaking
- one runner gradually gains a small lead during the shot

Emotion and performance
Their faces should show competition-level effort and determination:
- focused eyes
- tension in the jaw
- concentrated breathing
- effort visible in facial muscles
- no relaxed smiling
- subtle changes of expression during the shot as the competition intensifies
- occasional quick side glances at each other
- the feeling should be: "everyone wants to win"
As the shot progresses, let the competitive situation become more visible:
- one athlete edges slightly ahead
- others react and push harder
- expressions become more intense
- the group remains tightly competitive

Realism and motion quality
The video must be maximally realistic:
- realistic human anatomy
- realistic foot strikes
- realistic stride cycles
- natural arm movement
- natural hair movement
- realistic clothing movement
- natural skin texture
- correct lighting and shadows
- correct scale and perspective
Very important:
- no smear
- no melting
- no ghosting
- no extra limbs
- no flickering faces
- no identity drift
- no outfit changes
- no logo distortion
- no blurred hands or feet
- only very subtle physically natural motion blur, not AI blur

Style target
Premium sports-commercial realism, like a high-end athletic brand ad shot on a stabilized tracking camera in daylight.

Negative prompt:
blurry motion, smear, ghosting, double bodies, extra arms, extra legs, warped faces, changing identities, changing outfits, distorted logos, broken anatomy, floating feet, sliding runners, camera shake, sudden zoom, lens warping, flicker, jelly motion, unnatural body deformation, oversmoothed skin, cartoon look, fake background, inconsistent street geometry, duplicated runners, casual smiling jog, low-detail faces

Locked-off static street-level shot looking down Burlingame Avenue on a bright midday: blue "Burlingame" banners on black lamp posts, Peet's Coffee and its cafe tables on the right, parked cars on the left, shoppers on both pavements. The camera never moves. Six women runners begin about 20 metres away, small in frame, spread unevenly across the width of the road, and sprint straight toward the camera at a steady realistic pace, growing larger smoothly with no jump in distance, arms pumping, long driving strides, intense concentrated faces. The woman in the light blue patterned STRONGER set and the woman in the navy STRONGER shirt lead at the front; the black Nike set, the white Adidas top with green track pants, the pink HOKA set and the white and grey New Balance outfit follow close behind at different distances. Faces, hair, outfits and printed logos stay sharp and consistent. Buildings, banners, lamp posts, cars, tables and passers-by stay exactly where they are. Sound: trainers striking stone paving, getting louder as the group closes in, hard rhythmic breathing, faint street ambience. No music.

For the target video, at 0.00 seconds into the target video, <Picture 1> (from [Shot 1]) is fully referenced.

integrated_multimodal_description: [Shot 1] Live-action, photorealistic footage of a premium sportswear ad filmed on location, 16:9 landscape at real-time speed, one continuous take that opens on <Picture 1>: six women race hard down the middle of the sunlit shopping street toward the camera, preserving the street, storefronts, parked cars, street trees, the late sun and every runner's face, hair, outfit and position. The lead runner on the double yellow line wears the light-blue strawberry-print sports bra and high-waist leggings with two white stripes above the knee; a stride behind her, on the left of the frame, runs the copper-haired woman in the washed navy oversized t-shirt reading "STRONGER". The camera sits on a stabilised crane arm mounted on a camera car driving ahead of the pack, so every move is smooth and steady, and its path is one continuous movement. Until 00:01.500 it tracks backward in front of the pack in the wide view of <Picture 1>. From 00:01.500 the crane arm swings in an arc shot around the lead runner's left side, away from the woman in the STRONGER t-shirt, while it pedestals down with large amplitude and the zoom lens narrows, so that by 00:03.000 the camera runs alongside the lead at knee height and her white running shoes strike the asphalt across the lower frame, the strawberry-print leggings and white knee stripes flexing with every stride. From 00:03.000 the arm pedestals up at fast speed along her body to shoulder height and settles into a tracking shot level with her: a long-lens side profile framing the lead runner from the chest up, with the copper-haired woman in the "STRONGER" t-shirt pulling level just beyond her shoulder, both faces large in frame and in sharp focus, while the other four runners and the storefronts behind them fall softly out of focus. All six run at a steady hard racing pace of about three steps per second from the first frame to the last: long strides, forward lean, knees driving forward, heels kicking up behind, arms pumping hard, feet striking the asphalt flat, strides out of step with one another. Their faces show real effort with natural skin texture, visible pores and a light sheen of sweat: mouths open, breathing hard, jaws set, eyes fixed ahead. Warm, low late-afternoon sun from the far end of the street behind the runners rim-lights their hair, shoulders and arms and lays long shadows across the asphalt, while a large soft warm bounce from the camera side keeps every face evenly lit; the exposure stays constant through the whole move. The street trees show individually defined leaves stirring slightly in the breeze, the asphalt shows fine grain, small cracks and dark tar seams, and the brick sidewalk keeps crisp individual bricks. Natural motion blur on hands and feet from a 180-degree shutter at 24 fps, fine film grain and one consistent warm grade. The video ends at 00:05.170 with both women still at full racing speed in side profile.

overall_soundscape: Sneakers slap the asphalt in an uneven, overlapping rhythm under the low hum of the camera car. As the camera drops beside the lead, her shoe impacts turn sharp and close, with grit crunching under the soles, and as it rises to her shoulder her hard, rhythmic breathing and the flutter of the STRONGER t-shirt come forward. Distant birds and a faint city hum sit underneath throughout.

non_diegetic_music: An electronic track at about 176 beats per minute with a tight kick drum and clipped hi-hats landing on the footfalls, and a synth bass line that grows louder as the camera rises to the final profile.

Live-action, photorealistic footage of a road race, 5 seconds, real-time speed, continuing directly from the first frame. Six women race hard down the middle of a sunlit shopping street toward the camera in the last kilometre of a 10 km race.

SCENE, exactly as in the first frame: a closed two-lane street with a double yellow centre line, brick sidewalks, storefronts with grey metal awnings and hanging flower baskets on black lamp posts on the left, a white storefront with a black sign on the right, parked cars along both kerbs, street trees on both sides and a low beige building with an arch at the far end. The road ahead of the runners is clear.

RUNNERS. Each woman keeps the exact face, hair, body, outfit and shoes she has in the first frame and stays in her own lane for the whole shot. Faces stay sharp and detailed with natural skin texture:
- Lead, nearest, on the double yellow line: brunette with a slicked-back ponytail in the light blue strawberry-print sports bra and matching high-waist leggings with two white stripes above the knee, white running shoes.
- A stride behind her on the left of the frame: copper-haired woman in a washed navy oversized t-shirt with white arched letters reading "STRONGER", navy bike shorts and white socks with navy stripes. The letters stay crisp and readable as the fabric moves.
- In the left lane beside the parked cars: woman with a curly puff in a mauve HOKA crop top and dusty pink running shorts.
- Far left on the brick sidewalk: woman with a long black ponytail in a black Nike sports bra and black leggings.
- Right of centre, further back: ash-blonde woman in a white sports bra and sage green track pants with white side stripes.
- Far right, furthest back, beside a white SUV: dark-haired woman in an open light grey zip jacket over a white top and black shorts.

MOTION. From the very first frame all six run at a steady hard racing pace of about 18 km/h, three steps per second, and that pace holds unchanged to the last frame: long powerful strides, forward lean from the ankles, knees driving forward, trailing heels kicking up high behind, arms pumping hard from hip to chin with elbows at ninety degrees, strides out of step with one another, every planted shoe landing flat on the asphalt and pushing off.

CAMERA. One continuous tracking shot from a camera car driving backward down the centre of the road just ahead of the lead runner, lens at chest height, wide 28mm lens, deep focus so the street stays detailed. The camera vibrates slightly with small amplitude from the road surface. The lead runner is slightly faster than the car and gains about one metre over the shot, while parked cars, tree trunks and lamp posts slide backward past both edges of the frame at about one car length per second.

BEATS
0:00-1:30 already at full speed: the lead's left foot strikes the asphalt and meets her long shadow as her right heel kicks up behind her.
1:30-3:00 the runners pass under a street tree: dappled leaf shadows slide over the lead's shoulders, face and arms, then she runs back into warm sunlight. The copper-haired woman in the STRONGER t-shirt pushes harder and closes half a stride, exhaling through an open mouth, the loose t-shirt rippling.
3:00-4:20 the lead exhales sharply and drives her arms harder, her ponytail whipping side to side; the grey jacket of the furthest runner flies open behind her.
4:20-5:00 still at full racing speed in the final frame, the lead mid-stride with both feet off the ground.

FACES AND BODIES. Real effort: mouths open and breathing hard, jaws set, eyes on the road ahead, natural blinks, flushed cheeks, sweat shining on foreheads and shoulders, loose strands of hair on damp skin. Thigh and calf muscles tighten on each foot strike; the strawberry print and white stripes stay locked to the lead's leggings.

LIGHT. Warm, low late-afternoon sun from the far end of the street behind the runners is the only light source: it rims hair, shoulders and arms, long shadows stretch from their feet toward the camera and slightly left and travel with them, faces sit in soft shade lit by bounce from the pale asphalt. The sun stays in place and exposure stays constant.

CAPTURE. Real footage at 24 fps with natural motion blur on pumping hands and flying feet from a 180-degree shutter, fine grain, true-to-life colours, one continuous shot.

SOUND. Rubber soles slapping the asphalt in an uneven overlapping rhythm, hard breathing from the two nearest runners, the camera car's low engine hum and wind on the microphone, birds in the street trees. The soundtrack is only the street, the car and the runners.

@Image1 is the first frame: the street, camera position, sunlight and each runner's place at 0.0s. @Image2 is identity and wardrobe only, top row left to right: 1 light blue strawberry-print STRONGER set, 2 washed navy oversized STRONGER t-shirt and bike shorts, 3 black Nike, 4 white adidas top and sage track pants, 5 mauve HOKA top and pink shorts, 6 open grey New Balance jacket. Take only the six women from @Image2, never its grey backdrop, flat light or lineup. Preserve each woman's exact likeness and outfit throughout, no beautification.

16:9, 5 seconds, real-time speed. Live-action sportswear ad filmed on location with a stabilised gimbal on a camera car, 24 fps with natural motion blur, fine film grain, gentle vignette, one warm grade. Palette: warm gold sunlight, cool blue-grey shade, pale asphalt, the sky-blue and navy STRONGER outfits the brightest colours in frame.

Light as in @Image1: the only light is the late-afternoon sun, low at the far end of the street and slightly right, about 15 degrees high, behind the runners. Hard warm sunlight with crisp shadows; the shade stays cool. A bright warm rim on hair, shoulders, arms and calves; long shadows fall toward the camera and slightly left. Faces sit on the shadow side, one to two stops darker, lit by soft warm bounce from the asphalt, with real skin texture and pores and a soft glow where the backlight wraps their edges. From the side the same sun comes from frame right, behind them, and their shadows stretch ahead. The sun never moves and exposure stays constant; dust kicked up by their shoes catches the backlight.

Breathing: every runner takes a quick inhale with lips nearly closed, then a hard exhale through an open mouth, every second step, so mouths open and close in rhythm.

[0:00-1:50] Camera: from the front view of @Image1 on a wide lens, the gimbal swings around to frame right, the runners' left side, dropping to chest height and closing to three metres. The six women race hard toward it at full speed.
[1:50-2:50] Camera: tracks parallel beside them, slightly slower so they pull forward, the zoom narrowing so the four runners behind go soft. The lead in the strawberry-print set runs closest, half the frame height, the sun rimming her ponytail. The copper-haired runner in the navy STRONGER t-shirt surges to her shoulder, lips pressed, then a sharp exhale.
[2:50-3:50] Camera: keeps tracking as shopfronts, trees and parked cars never in the opening frame sweep past, over two frame widths, leaf shadows sliding across the runners. Seeing the rival at her shoulder, the lead's brow tightens and her jaw sets.
[3:50-5:00] Camera: holds the side track past a new corner shopfront. The two STRONGER runners run level, stride for stride, eyes narrowed, nostrils flaring on each inhale.

Motion: long strides, fast cadence, arms driving hard, ponytails swinging with each stride, the loose t-shirt hem lifting and slapping flat, logos readable, heavy speed blur across the street while the runners stay sharp.

Sound: rubber soles slapping asphalt in an uneven rhythm, hard breathing with each exhale matching an open mouth, t-shirt fabric fluttering, soft street ambience, no music.

Stable faces throughout, realistic physics, no morphing.

runnig6_street_race_16x9_v3 is the first frame of the video: the street, camera position, sunlight and the position of every runner at 0.0s.

six_women_sheet_2 is the identity and wardrobe reference only: the faces, hair, bodies, outfits and shoes of the same six women, top row left to right: 1 the brunette with a slicked-back ponytail in the light blue strawberry-print STRONGER sports bra and leggings, 2 the copper-haired woman in the washed navy oversized STRONGER t-shirt and bike shorts, 3 the woman in black Nike, 4 the ash-blonde in the white adidas top and sage green track pants, 5 the woman in the mauve HOKA top and pink shorts, 6 the woman in the open grey New Balance jacket. Take only the people from six_women_sheet_2; its grey backdrop, flat studio light and side-by-side lineup stay out of the video. Every woman keeps her exact face and outfit from six_women_sheet_2 throughout, with a stable, recognisable face. 16:9.

Side-tracking shot down a shopping street. Light, exactly as in runnig6_street_race_16x9_v3: the only light is the late-afternoon sun, low in the sky at the far end of the street and slightly to the right, about 15 degrees above the horizon behind the runners. It is hard, direct sunlight with crisp shadows, warm golden in colour, while the shade stays cool and slightly blue. It rims the runners' hair, shoulders, arms and calves with a bright warm edge that separates them from the street, and throws long hard shadows from their feet toward the camera and slightly left. The fronts of their bodies and their faces are on the shadow side, one to two stops darker, lit only by soft warm bounce from the pale asphalt, with natural skin micro-texture and subtle pores. When the camera reaches their side, the same sun comes from the right edge of the frame, behind them: the backs of their heads, shoulders and calves glow, their faces turned toward the left of the frame stay in soft bounce light, and their shadows stretch ahead of them across the asphalt. Fine dust kicked up by their shoes catches the backlight. The sun does not move and the exposure does not change during the camera move; only the dappled shadows of the street trees slide over the runners as they pass under them.

The video starts exactly on @Image1: six women racing hard straight toward the camera down the middle of the street. Over the first 1.5 seconds the camera swings from this front view around to the right side of the frame, the runners' left side, dropping to chest height and moving in close, then travels parallel with them for the rest of the shot, moving with the group but slightly slower so they pull forward through the frame. They cover about twenty-five metres of the street, so the background travels more than two full frame widths: the storefronts, lamp posts with hanging flower baskets and parked cars from the opening frame leave the edge within two seconds, and three or four shopfronts, street trees and parked cars that were never in the opening frame sweep in and pass behind them.

Breathing and faces: every runner breathes in a real running rhythm synced to her stride, a quick inhale with lips nearly closed, then a hard exhale through an open mouth, every second step, so mouths keep opening and closing instead of staying open. Their expressions change as the race tightens: eyebrows pull together, jaws clench, lips press into a line before a push, cheeks puff on a sharp exhale, eyes flick sideways at the rival next to them.

[0:00-1:50] The pack runs toward the camera as in runnig6_street_race_16x9_v3, backlit, while the camera swings to their side. The lead in the strawberry-print STRONGER set breathes steadily, face focused, lips closing between exhales.
[1:50-2:50] The camera is beside them; the lead runs closest, filling half the frame height, the sun rimming the back of her ponytail and shoulders. The copper-haired runner in the navy STRONGER t-shirt glances at her, presses her lips together, then surges up to her shoulder with a sharp exhale through gritted teeth.
[2:50-3:50] They pass under a street tree and leaf shadows slide across their faces and arms, then warm sun hits their backs again. The lead feels the rival there: her brow tightens, her jaw sets, and she drives her arms harder with a short, strong exhale. Behind them the runner in the mauve HOKA top and the blonde in green track pants push hard, faces straining, mouths opening wide on each breath.
[3:50-5:00] The two STRONGER runners are level, stride for stride, passing a new corner shopfront; both faces show the last extra effort, eyes narrowed, nostrils flaring on the inhale, mouths snapping open on each exhale. Full racing speed, fast leg cadence, arms driving hard, ponytails and hair whipping backward and glowing in the backlight, faces sharp, clothing and printed logos readable, heavy speed blur across the buildings, parked cars and pavement while the runners stay sharp.