SYSTEM NOTICE

Auto translation by AI. Be sure, accuracy, nuances and authorial intent may not be fully reflected.
見出し画像

7 Recommended Anima Workflows

Hello, this is Fukashigi-chan (´▽`)

In this article, I will share the ComfyUI workflows for Anima that I have been using recently. The workflows I actually use are quite messy, so this time I have divided them into seven functional categories to make them easier to use.


1. Basic Anima Workflow

You can generate images by typing in prompts.

[Updated 7/20] I now use a simpler workflow as shown below, using Anything Everywhere and SDXL Empty Latent Size Picker.

2. i2i (Image-to-Image) with Anima

You can create images based on existing ones. The lower the denoise strength in the KSampler, the closer the result will be to the original image; the higher it is, the more it will deviate from the original.

3. Inpainting with Anima

Use this when you want to modify only a part of an image. Once you load the image, right-click on it and select "Open in MaskEditor" near the bottom to enter the mask editing screen, then simply paint a mask over the area you want to modify.

4. Outpainting with Anima

Use this when you want to add content to areas outside the original image. I often use this when the top of the head or the tips of the hands are cut off. Load the image and enter numbers in the direction you want to extend it using the "Pad image for outpainting" node. (For example, if you want to extend the top by 100 pixels, set the top value to 100.)

5. Anytest with Anima

This is a tool that can do various things, such as creating illustrations from line art. It can do so much that it's hard to explain, so please try using it in various ways or look up and read specialized articles. I often use it by inputting images created with NovelAI.

Input an image like this...
And you can create an image like this.

6. The Recently Popular Regional-Controlnet

This is something that might be useful when generating multiple characters in a single image. From what I've tried, the elements tend to get mixed up quite a bit, so you'll likely end up needing to roll the dice or fix it with inpainting. Using the "Open in MaskEditor" mentioned above, select the brush icon on the left and paint the areas where you want to place the characters. For now, I have set it up so that prompts are reflected in the areas painted in red, blue, and yellow.

Paint with this
If you paint like this...
You can create an image like this.

⑦ ADetailer in Anima

You can use ADetailer, which was very popular in Stable Diffusion, in Anima. Since eyes can sometimes get crushed in full-body illustrations, using this to fix them can often make them look beautiful.

Please leave a comment if you have any questions.

Please search for the necessary nodes and models yourself or ask ChatGPT. If you throw the JSON file directly into ChatGPT, it should be able to tell you various things.

If you really don't understand, please write it in the comments and I will answer (´▽`)

いいなと思ったら応援しよう!

不可思議ちゃん@AI漫画でkindle出版 😄👍✨

この記事が参加している募集