10s 9:16 Vertical Five-Shot Transition Video AI Image Prompt

Generate a 10s 9:16 vertical video using the uploaded character image and audio beat markers: Shot 1 (0-2s): First-person wide shot, under a clear blue sky, an azure salt lake and emerald grassland mirror-like. The character has long hair draped over shoulders, wearing floral accessories, a white flowing backless halter dress, smiling while holding hands, turning and spinning under the arm for transition. Shot 2 (2-4s): Smooth transition to in front of plateau snow mountains (blue sky, snow mountains, green land, streams, cattle and sheep), first-person wide shot. The character changes into a green plant print halter dress and a woven straw hat with flowers, holding hands and turning to spin for transition. Shot 3 (4-6s): Transition to under a colorful prayer flag dome (sunlight filtering through prayer flags onto a small pool), the character changes to long wavy hair, white halter top + blue knitted shawl + turquoise belt, holding hands and spinning by the pool for transition. Shot 4 (6-8s): Transition to desert dunes (azure sky, a lying camel on the left rear), first-person wide shot. The character changes into a blue gold-trimmed belly-baring exotic outfit, hair in a bun, wearing metal tassel headdress and hand ornaments, facing the sun, holding hands and turning to spin. Shot 5 (8-10s): Transition to a colorful mineral lake (yellow, orange, blue interwoven lake water, distant mountains and white clouds), the character wears a red dress and a bun, holding hands and spinning, the scene ends in a hand-holding state.