【🦜Midjourney以图生图】最新超详细新手入门教程！独家技巧公开！教你保持人物统一连续，用ChatGPT+AI绘画软件制作高质量故事短片赚取收入，怎么写提示词，垫图与权重参数—第4集｜暗夜飞行

The video script introduces a popular method of using AI tools like ChatGPT for generating story plots and Midjourney or DALL-E for creating images. It explains that while the process seems simple, maintaining character consistency across different images is challenging. The video aims to demonstrate how to maintain character continuity in images, a technique endorsed by Midjourney. It covers advanced techniques such as 'padding' (using an existing image to generate a new one) and the use of 'weights' to control the influence of the initial image on the final output. The script also provides a step-by-step guide on how to use these techniques, starting with generating a story plot from ChatGPT, segmenting the story into scenes, and then using Midjourney to generate images that maintain the main character's continuity across different scenes.

This paragraph delves into the technical aspects of the image generation process using Midjourney. It explains the concept of 'Image Weight' (IW), which controls the influence of the initial image on the final result, with a range from 0 to 2. The paragraph uses an example to illustrate how increasing the IW value shifts the generated image's focus from a text description ('birthday cake') to the initial image ('flower'). It also introduces 'multi-prompts', denoted by double colons, which serve as separators to ensure that different elements in the prompt are treated as independent, thus preventing confusion and maintaining clarity in the generated images. The paragraph concludes with a practical example of how to apply these concepts to generate a series of images that tell a cohesive story.

The final paragraph emphasizes the iterative nature of AI image generation. It highlights that achieving the desired outcome often requires patience and multiple attempts. The script shares the creator's experience of generating over a hundred images to select the eight most satisfactory ones for the story. It also addresses the need for fine-tuning and adjusting the process to meet specific requirements, such as adding a missing character (a small boy) to the scene through a secondary prompt. The video concludes with an invitation to view the complete fairy tale story narrated by the creator, encouraging viewers to maintain their sense of wonder and to support the channel by liking and following for more quality content.