Continuous Descent from Stratosphere to Coastal City Core AI Image Prompt

1. First-frame lock (Image-to-Video baseline) Use the reference image as the only first frame, strictly preserving the original camera height, wide-angle spherical perspective, geographic forms, and colors: ultra-high-altitude overhead view of a coastal metropolis; the upper edge of the frame shows a clear Earth curvature and an atmosphere gradient from deep blue to light; the midground is a turquoise shallow bay wrapped by a crescent-shaped golden beach, with several cargo ships, container ships, and small speedboats moored and sailing in the bay, white wake trails clear; on the right is a dense cluster of glass-curtain-wall skyscrapers and extending city blocks, interwoven with dark green mountains and hills; in the distant left is a cross-sea bridge, port docks, and container yards; the sky is rich cobalt blue, scattered with volumetric cumulus and cirrus clouds; large foreground white clouds obscure the lower edge and corners of the frame. 2. Overall style anchor (unchanged throughout) Epic high-altitude aerial / documentary cinematic feel / stream-of-consciousness grand narrative; main colors are cobalt blue—turquoise—sand gold—deep green—concrete gray-white; noon natural sunlight, high-altitude clear transparency, cloud shadows naturally cast on ground and sea; ultra-wide-angle slight spherical distortion, ultra-high-resolution surface details, realistic atmospheric perspective and distant blue haze, cinematic latitude and slight film grain; stable, slow, solemn camera movement (drone/high-altitude platform quality, with very slight airflow float); throughout only natural light and real weather movement, no special effects, light effects, particles, lens flares, text subtitles, UI interfaces, or dialogue. 3. Storyboard script (8 shots, continuous story: one continuous descent flight from stratosphere to city core) Shot 1 | 0.0–1.2s | Critical altitude (slight first-frame motion) From the first frame, maintain ultra-high-altitude viewpoint; the camera begins with extremely slow forward drift; Earth curvature and atmosphere gradient remain stable, high-level cirrus drifts slowly laterally, white wake trails from ships on the sea extend very slowly, the city maintains a static macroscopic texture. Shot 2 | 1.2–2.4s | Descent begins The camera begins extremely slow vertical descent while pushing in (descend + push-in); foreground white clouds at the lower edge sweep out of frame from bottom to top; the bay and beach outlines gradually enlarge in frame; mountain textures and block grids begin to appear. Shot 3 | 2.4–3.6s | Through clouds The camera continues descending through a thin cloud layer; filamentous cloud wisps rapidly sweep past the foreground (natural cloud pass-through, not special effects); through cloud gaps the crescent beach and turquoise shallow sea become clearer layer by layer; dark and light patches of underwater reefs become distinguishable. Shot 4 | 3.6–4.8s | Bay shipping lane The camera looks down and slowly pans across the bay; cargo ships move forward at real slow speed, wakes spread in a V shape; a small speedboat cuts a slender white line across the bay mouth; cloud shadows slowly sweep across the sea surface causing light-dark transitions. Shot 5 | 4.8–6.0s | Coastline The camera slowly tracks laterally along the crescent beach; fine white wave lines push up onto the beach layer by layer and retreat; dense traffic on the coastal avenue moves slowly like a thin stream; seaside green belts and pier details are clear. Shot 6 | 6.0–7.2s | Above the building cluster The camera continues descending and flying forward to directly above the CBD skyscraper cluster; glass curtain walls reflect the moving sky and clouds (natural reflection); long building shadows slowly cast on blocks; rooftop equipment and intersection traffic details are visible. Shot 7 | 7.2–8.5s | Port and bridge The camera makes a slight slow yaw toward the cross-sea bridge and port on the left; bridge traffic forms lines; shore cranes and container yards are neatly arranged; a giant ship is slowly docking; background mountain ridges layer in atmospheric blue haze. Shot 8 | 8.5–10.0s | Descending toward city core (ending) The camera returns to the main axis, continues slowly descending and advancing toward the city center; the field of view gradually tightens between tall buildings; bay, beach, and building cluster are compressed into depth layers in the same frame; cloud shadows, ships, and traffic still flow; the ending keeps the descent and push-in momentum and steadily concludes, without pulling out, climbing, or cutting to black. 4. Consistency and technical requirements Full-film geographic structure completely consistent: bay shape, beach arc, building distribution, mountain orientation, bridge and port positions unchanged; no adding/removing buildings and islands Maintain the first frame's ultra-wide-angle spherical perspective and Earth curvature feel; distortion transitions naturally during descent; do not switch to telephoto eye-level Light direction, cloud forms, and color temperature unified throughout; cloud shadow movement direction consistent with cloud drift direction All motion conforms to real high-altitude scale speed: slow ships, gentle waves, fine and slow traffic, slow cloud drift; no accelerated fast-forward Transitions only use continuous camera movement or natural slow cuts; no jump cuts, flash cuts, camera roll, or frame shake 24fps cinematic feel, 9:16 vertical, 8K surface details, realistic atmospheric perspective and slight grain Sound effects (optional, no voice): high-altitude wind, low-frequency air rumble, very distant sea waves and city ambience 5. Negative Prompt Light effects, lens flares, glowing particles, scan lines, sci-fi HUD interfaces, data UI, text subtitles, watermarks, logos, coordinate annotations, dialogue voices, time-lapse accelerated feel, traffic light trails, night neon, sudden color shifts, sudden weather changes, rain lightning, terrain distortion deformation, melting/repeating buildings, drifting misaligned ships, excessive fisheye distortion, frame flicker shake, fast rotating camera, ending pull-out/zoom-out/climb, cartoon rendering, fake HDR oversaturation.