Dream Land Video Breakdown (Prompts Included)
Follow this two-step workflow to turn a character concept and surreal setting into a cinematic AI video with Pollo AI. Start by creating the reference images, then animate them with your preferred video model.
Step 1: Image Generation
Open Pollo AI and select Image Generation. Paste the prompt below, then choose GPT Image 2 or Nano Banana 2.
Reference Images
Use the prompts below to create one character reference image and one scene reference image.
Screenshot Example
Create Your Dream Land Video on Pollo AI
Create dreamlike characters and surreal scenes with AI, then bring them together in one cinematic video.
Try Pollo AI for Free
Example Prompt
Create a professional four-view character reference sheet featuring the same 23-year-old Western European female rap-pop performer with an early-2000s Y2K aesthetic and K-pop-level polish.
Layout: Use a 16:9 canvas with four left-to-right views: 1) a head-and-shoulders beauty close-up occupying 30% of the width, 2) a front full-body view, 3) a 90-degree left-side full-body view, and 4) a back full-body view. Keep the three full-body panels equal in width, with consistent character height, a shared floor line, minimal perspective distortion, and an orthographic fashion-turnaround style.
Character: 170 cm tall with a slim adult build, Western European facial features, deep-set blue-gray eyes, fair skin with cool pink undertones, subtle freckles across the nose and cheeks, and realistic skin texture for a 22- to 24-year-old. Do not resemble any celebrity.
Image Prompt (Character)
Prompt
A realistic close-up portrait of a young adult woman with natural strawberry-blonde hair, fair skin, subtle freckles across her nose and cheeks, green eyes, and naturally full pink lips. She looks directly at the camera with a calm, neutral expression. Her loose, voluminous hair frames her face and blows gently to the sides. She wears an oversized dark forest-green college hoodie with partial white varsity lettering. High-angle selfie framing, tightly cropped face, shoulders and hoodie filling the lower frame, softly blurred suburban street background, bright afternoon sunlight, crisp shadows, natural skin texture, candid smartphone photography, realistic colors, warm golden hair tones, dark asphalt background, square composition, shallow depth of field, high detail, natural beauty photography.
Negative Prompt
Heavy makeup, glamour studio lighting, plastic skin, airbrushed face, exaggerated eyes, distorted face, elderly or childlike features, harsh beauty filters, low resolution, blur, artificial hair color, text, logos, watermarks, extra fingers, unnatural poses, overexposed skin.
Image Prompt (Scene)
Prompt
A surreal, dreamlike architectural garden in a square composition. A grand pastel-pink staircase curves upward across the center-right, with tiny figures on the steps for scale. Lush Mediterranean-style gardens surround both sides, filled with dense green foliage, red flowers, and blue-purple blossoms. Reflective water flows beneath the staircase, while massive polished chrome spheres float among classical columns and arches. At the center, a mirror-like liquid hourglass reflects the sky and garden. Bright blue sky, soft cumulus clouds, sparkling starbursts, lens flares across metal surfaces and stair edges, warm sunlight, glossy reflections, surrealist subconscious imagery, magical realism, retro-futurist 3D rendering, cinematic depth, intricate detail, soft yet vivid colors, peach pink, sky blue, emerald green, chrome silver, dreamy atmospheric haze.
Negative Prompt
Low resolution, blur, flat lighting, dull colors, modern cities, ordinary realistic stairs, close-up people, text, logos, watermarks, distorted architecture, cluttered composition, dark mood, monochrome, horror, cyberpunk.
Step 2: Video Generation
Upload the generated images, select Video Generation, and choose Reference to Video. Select Seedance 2.5, Seedance 2.0, MiniMax H3, or Wan 3.0, paste the prompt below, and click Generate.
Screenshot Example
The example below shows the generated reference image uploaded in Reference to Video, with Seedance 2.5 selected.
Video Prompt
Global Direction
Use the reference video only as a guide for its dreamy cinematic lighting, ultra-wide-angle framing, seamless single-take camera movement, fluid physics, surreal spatial design, and hypnotic dreamcore soundtrack. Do not include ambient foley such as wind or train noise, specific brands, or text. 3:4 aspect ratio, 24 fps, 15 seconds.
0-3s
Begin with a continuous wide-angle POV shot looking through an open cyan train door. The train glides past water covered with colorful wildflowers and floating chrome spheres that reflect the sky and light. Dreamy music plays throughout, featuring lush synth pads, soft hypnotic guitar, and floating ambient chimes, with no realistic sound effects.
4-8s
Without cutting, rotate the ultra-wide camera to reveal a young woman with long, wavy reddish-blonde hair and freckles. She wears a white T-shirt and light-blue jeans and leans through the train door while holding a silver bar. Her hair streams in the wind against sunlit waterfalls and floating islands as the music develops into a warm, enchanting rhythm.
9-12s
Glide closer to capture the moving train reflected in a massive floating mirrored sphere alongside the woman’s sunlit, awe-filled expression. Natural lens flares appear as petals drift through the air and the music rises into a soaring, surreal chorus.
13-15s
Continue the same take as the camera pulls back into a wide cinematic view. The cyan train speeds across liquid tracks toward a towering futuristic spire beneath the sunset sky. The music resolves into a soft, lingering final chord with no environmental noise.
Final Summary
Create a 15-second, 3:4 dreamcore train sequence in one seamless ultra-wide take. Keep the character and surreal setting consistent, use only the specified music, and avoid cuts, ambient foley, brands, and on-screen text.
Want to bring your Dream Land video to life in minutes? Upload your character and scene references, paste the prompt, and create your own cinematic dreamworld with Pollo AI.