Original document
GPT Image 2 · en
Create an emotionally resonant banner poster for a Japanese movie trailer, employing a high vertical composition and a cinematic anime illustration style. The background is set on a Japanese city street on a rainy night, the ground reflecting the wet asphalt, neon signs lining the streets, and distant skyscrapers obscured by the haze. In the foreground, the silhouette of a young anime girl is shown, facing left, standing alone under a transparent plastic umbrella, raindrops visible on its surface. She has long, dark purple hair tied in a high ponytail, her expression slightly melancholic, wearing an oversized light blue hoodie and a loose pink-blue plaid coat. The overall atmosphere is lonely, melancholic, and dramatic, like a live-action film reimagined as an anime key visual. Use cool blue and cyan tones, accented with pink neon lights, to create a glossy rain reflection, a shallow depth of field, a soft, atmospheric haze, and subtle contour lighting from the street signs. Add a striking handwritten white Japanese title at the top.{argument name="title text" default="AI's Tears"} Add a smaller white subheading below.{argument name="tagline text" default="Why did she cry?"} Add a line of bold white release date at the bottom.{argument name="release text" default="Coming to theaters nationwide in Fall 2024"} Maintaining the character as the emotional focus, the composition employs a half-body shot from the waist up, with the umbrella arching above the head. The empty, rain-soaked street in the background further emphasizes the sense of loneliness and the cinematic atmosphere of sadness.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.
