Create a 5-second vertical cinematic fantasy adventure… | Deepfake AI视频

生成此视频所用的文字转视频提示词

Create a 5-second vertical cinematic fantasy adventure video based directly on the reference image. SUBJECT: A fluffy white cat and a golden retriever. Preserve their exact appearance, fur colors, body proportions, relative size, and identities from the reference image. ENVIRONMENT: A magnificent magical city of books floating above a sea of clouds. The animals are walking on the pages of an enormous open book. Along the floating street are bookshops, towering stacks of books, ancient architectural towers, tiny distant townspeople, and countless loose pages drifting through the bright blue sky. A large hanging banner reads "THE CITY OF BOOKS". CAMERA: Match the original rear three-quarter view of the animals walking toward the city. Keep both animals visible in the foreground and preserve the floating city and book market in the background. Use a smooth, gentle forward tracking movement. Maintain the original composition and sense of scale. One continuous shot, no cuts. TIMELINE: 0–1 seconds: The white cat and golden retriever continue walking together across the enormous open book toward the magical book market. The floating pages drift gently through the sunlight. The distant townspeople continue their normal activities. 1–2 seconds: A single loose parchment page descends from above and gently lands on the giant book directly in front of the white cat. The cat notices the page and stops walking. The golden retriever takes one more small step before noticing that the cat has stopped. 2–3 seconds: The white cat lowers its head and gently taps the parchment page once with one front paw. The page slides slightly across the paper surface. The cat watches it with focused curiosity. 3–4 seconds: A gentle gust of magical wind lifts the same parchment page from the book. The page rises into the air and begins floating forward along the street toward the distant book market. 4–5 seconds: The white cat follows the floating page with its eyes, then begins walking after it. The golden retriever immediately follows the cat. Both animals continue together toward the heart of the magical city. End with the two animals following the drifting page into the city. VISUAL STYLE: Photorealistic cinematic fantasy animation, realistic fluffy fur, natural animal movement, detailed aged paper, enormous open books, intricate bookshop displays, ancient floating towers, warm golden sunlight, luminous clouds, delicate atmospheric haze, and a strong sense of magical wonder. The scene should feel bright, inviting, and full of discovery. The magical event is subtle: one ordinary-looking parchment page moves under the influence of an invisible magical breeze. Do not turn the page into a creature. AUDIO: Gentle wind above the clouds, soft paw steps on thick paper, distant bustling book-market ambience, and delicate paper rustling as the parchment lands. A soft fluttering sound accompanies the page lifting into the air. Add one subtle magical chime as it begins floating forward. No dialogue. The visual story must remain completely understandable even if no audio is generated. CONTINUITY AND CONSTRAINTS: The white cat and golden retriever remain the same animals throughout the video. Both animals remain on the giant book pages and never jump into open air. The same single parchment page remains visible from its arrival until it floats away. The page lands before the cat touches it. The cat touches the page only once. The page lifts only after the cat has touched it. The page floats forward along the street, away from the camera. The cat begins following the page before the dog. The dog follows the cat. The page does not transform, multiply, disappear, or become an animal. The giant book, bookshops, banner, towers, clouds, and distant buildings remain consistent with the reference image. The townspeople remain background figures and do not approach the animals. Do not introduce additional animals, magical creatures, or new characters. No sudden camera movement, teleportation, or scene changes. No cuts. The visual story must be clear without audio.