🟢 Prompt Auto-Generation

Recently, Midjourney announced support for the image-to-prompt feature.

Original Image

blueprint holographic design of futuristic Midlibrary --v 5

Prompt Generation

1. Directly input the /describe command.
2. Upload the image through the pop-up window and send it. Midjourney will generate four possible prompts based on the image.
3. Click the 1, 2, 3, or 4 button below the image to directly send the prompt.

technology research and industry background concept stock photo, in the style of fairy academia, illuminated interiors, hyper-realistic sci-fi, studyplace, light azure, bibliopunk, scientific diagrams --v 5

Advantages

With the /describe command, you can:

1. Learn new prompt strategies and expand MJ vocabulary: Midjourney primarily uses recognizable and intuitively interpretable words to describe images. Therefore, in most cases, if /describe uses a word, expression, or style modifier to describe an image, we can use it in our own prompts.
2. Explore new artists: Similarly, the artists that Midjourney "recognizes" in the input image can be valuable. Every name in the output prompt means (in most cases) that MJ recognizes that artist.
3. The best tool for Remix: You can see how AI reinterprets your visual art. You can view it from a new perspective or gain inspiration from new and unique ways to further develop your work. Turn on Remix mode before inputting /describe, so you can further adjust the prompt before sending it. Prompts generated by /describe cannot be used to create images identical to the uploaded image. /describe will return the aspect ratio of the uploaded image.

takeaways

/describe生成的提示不能用于生成与上传的图像一样的图。 /describe会返回上传图像的纵横比。