I tried to generate a high-quality image using the high-quality spell dictionary ``Magic Encyclopedia Dictionary'' that brings together the wisdom of image generation AI users
Image generation AI, which generates images according to instructions simply by entering text (prompts), is attracting a lot of attention. . A new web application ` `Magic Encyclopedia (Spell Encyclopedia) '' that allows you to create highly accurate prompts just by clicking on it by image generation AI enthusiasts in the Chinese-speaking world has been released, so I tried using it. rice field.
magic spell encyclopedia
The spell encyclopedia looks like this. By clicking the 'prompt fragment' displayed at the bottom of the screen, you can build a prompt that can generate your own image.
The Spell Encyclopedia is mainly composed of Chinese and English, which makes it difficult for Japanese users to use. In such a case, you can view the reference image by switching 'Display image (display image)' to ON.
The Spell Encyclopedia contains completed prompts as well as 'prompt fragments'. Click the button at the bottom of the prompt you care about, and the prompt will be displayed at the top of the screen and copied to your clipboard.
Below is an image generated with the illustration specialized model '
How to combine 'prompt fragment' is as follows. First, select a category on the right side of the screen and click the '+' button placed at the bottom of the 'Prompt Fragment' displayed on the left side of the screen. Then, the 'prompt fragments' selected on the right side of the screen are lined up.
Categories include `` background '', `` hairstyle '', `` clothes '', etc., and you can specify the characteristics of the image you want to generate in detail, such as `` cyberpunk background is good '' and `` black hair color is good ''.
When you have finished selecting the 'prompt fragment', click the red framed icon to copy the entire prompt to the clipboard. This time the prompt 'head out of frame,cyberpunk city,black_hair/black_eyes,sitting,arms behind head,tareme,casual wear,' was copied.
Here are the results generated by the prompt I created. As intended, I was able to generate an image of a ``black-haired casual girl sitting against a cyberpunk landscape''.
Although the enchantment encyclopedia is written in Chinese and English, some of the spellings, such as ``first person (first person, subjective perspective)'', ``rainy city (rainy city)'', and ``volleyball uniform (volleyball uniform)'' The words of can be read somehow. If it is still difficult, it is convenient to use the translation function that comes standard with Chrome.
Related Posts:
in Review, Web Application, Posted by log1o_hf