Original document
GPT Image 2 · en
Create a seamless 360-degree isometric cylindrical projection HDRI environment map showcasing a futuristic city's nighttime highway, with the perspective situated in the center of a wide, wet road beneath a dense network of modern urban viaducts. Employ a panoramic, wraparound composition with strong fisheye-style isometric cylindrical distortion at the top and bottom edges, designed specifically for Blender world backgrounds. The scene should include two large, curved elevated ramps, one on each side, arching towards the center, supported by multiple pillars and multi-level interchanges extending into the distance. The road is surrounded by a skyline of brightly lit office and apartment buildings, their windows rendered in cool blue and cyan, accented with touches of magenta and purple decorative lights. The road surface should be glossy and wet from rain, reflecting long strips of neon blue, purple, and pink city lights. Add subtle light trails and luminous reflections to the road surface to create a cyberpunk atmosphere. The sky should be a deep, cloudy, and textured night sky with faint lights at the zenith, and distant buildings surrounded by soft mist. The overall color scheme is dominated by deep indigo, electric blue, cyan, and purple. The image should have a cinematic, emotional, highly detailed, and photorealistic feel, with realistic city lighting and mirrored reflections on wet surfaces, making it suitable for use as an HDRI background. It includes slight panoramic seam artifacts and stretching distortion near the poles, giving it an AI-generated but usable environment map texture.How to use
- Copy the complete original prompt, including any parameters.
- Replace the subject, text and composition with your own requirements.
- Use a compatible external tool; check its current access conditions and review the result.
Before you start
- Check whether a reference image is required. A compatible model label does not mean this example was generated with that model.
- Keep names, figures and visible text accurate; results can vary between model versions.
