Text -> Image

AI 图像提示词模板

Documentary Style Image Generation with Nano Banana Pro

This prompt template is designed for Nano Banana Pro to generate images in the style of a documentary TV program, complete with split-screen scenes (2x2 grid), superimposed text (telop), and picture-in-picture (wipe). Users upload a person's photo and then use this prompt, customizing the documentary's theme to create different scenarios, such as a '24-hour ramen shop' or 'former idol working in a local factory.' - YouMind

推荐模型: gemini-3.1-flash-image-preview

关于这个 AI 图像提示词

This prompt template is designed for Nano Banana Pro to generate images in the style of a documentary TV program, complete with split-screen scenes (2x2 grid), superimposed text (telop), and picture-in-picture (wipe). Users upload a person's photo and then use this prompt, customizing the documentary's theme to create different scenarios, such as a '24-hour ramen shop' or 'former idol working in a local factory.' - YouMind

提示词

A cinematic, dutch angle, full body shot of a {argument name="subject" default="young woman"} as in <<<image_1>>> cooly walking in profile, one hand holding an off-white, {argument name="prop" default="mobile suit pilot helmet"} with metallic black visor. She wears a tight, detailed, {argument name="outfit" default="futuristic white mobile suit pilot suit"} with subtle signs of wear and tear, including scuffs and grime textures. Cooly gazing at the camera. The lighting is soft, cool, and diffused, creating a melancholic and introspective mood with a palette of pale blues, stark whites, and pops of muted grey. The background is a blurred, minimalist sci-fi interior with sterile, light-toned architecture. Shot on a 35mm lens with a shallow depth of field, focusing sharply on her face and body, high-quality digital photography, hyper-realistic, intricate details, soft cinematic highlights, ethereal atmosphere. Straight long hair parted in the middle hair style. Visor should have no reflection of the photographer

如何使用这个 AI 图像提示词模板

  1. 1复制提示词 —— 取走该模板的提示词与反向提示词。
  2. 2选择模型 —— 选一个推荐的 AI 模型以获得最佳匹配。
  3. 3生成 —— 一键打开工作台并生成你的作品。