AI Drawing Combat (3): Hand-drawn Mech Raging Man | Stable Diffusion Becomes a Productivity Tool for Designers

S: Did you install stable diffusion just to see the young lady?
I: Of course not, of course it is for the development of the company to find a way out~~

---

Learn in advance :

• Installation and problem solving reference: " Windows installation Stable Diffusion WebUI and problem solving records ";
• Problems when running and using " Sorting out various problems encountered when Windows uses Stable Diffusion ";
• Model application and parameters " Stable Diffusion Personally Recommended Various Models, Setting Parameters, Extended Applications, etc. Collection ";
• Prompt Words Create Spells " AI Drawing Prompt Words/Spells/Affixes/Keywords Usage Guide (Stable Diffusion Prompt Designer Operation Manual) ";
• Different types of models Models description " Analysis of different types of StableDiffusion model Models ";
• Draw character movements and details of hands and feet " Stable Diffusion Accurately draw character movements and details of hands and feet (ControlNet extension required) ";
• Comparison of various styles and practical application " AI drawing style comparison table / detailed research records and experience summary of drawing style samples ";

Can stable diffusion only be young ladies? No, today we're drawing a Mecha Rampage with stable diffusion.

The reason is that some great masters said that AI can't reach the level of human drawing. I don't accept it, because I really can't achieve it. . . . Haha, now I'm just using his sketches to render a Mecha Rampage.

1. Determine the skeleton of the character

• Identify character bones in openpose
  

• After the marking is completed, save the png and import it in the controlnet of the Vincent graph mode

• Use the edge control, otherwise the picture may only have Transformers~~
  The preview can let you know what the parameters look like is what you need.

2. Add prompt words

The value of cue word correlation (CFG Scale) can be increased to make the cue words as real as possible. I used it this time 25.

((mecha style)),8k, realistic, strong sharp, highlights, obvious refraction, ((mecha driver is wearing armor with an exposed head)), (short hair and bread), groin protector, (left arm with M134 Minigun)(right arm with M134 Minigun), (magazine is from M134 Minigun to giant iron), chest flash, a giant iron barrel on back, heavy mecha, wearing mecha, buildings, 

• Note that the weight of the controlnet should not be too low, otherwise the weight of the affixes in the prompt words may be stronger, resulting in deviations in the actions and details of the graph.

• A picture of 512 is produced, why is it this size? Because the resolution is higher, my current model is Transformers. . .

The parameters at that time were as follows, if

I wondered if the redrawing range was too high, and the repair sampling was too low, so I changed it to

The drawing is just not finished. . .

It may be a bug or some other places have not been adjusted properly, no matter, let's make a picture directly now.

3. Graphics

• Can only use direct scaling, otherwise I'm completely out of video memory.
  Remember to use controlnet to define the edges and bones when drawing a picture.
  

• Drawing

4. P map plus details

• Complete the ammunition barrels and bullet belts.
  

• Continue to generate pictures, let ai adjust it by itself

• After multiple graphs

I got a flu today. . . So there it is, in summary:

1. It is the best policy to have a foundation in art design. Unless you completely let AI give you creativity, the structure, layout, details, and colors you have conceived will take a long time to adjust before you can get close;
2. To understand the meaning of each parameter, you will get the ideal picture faster;
3. When the graphics card is not working, start with 512 pixels to see if your prompt words and parameters meet the requirements, and then use the larger image.
4. hardware hardware hardware~~

• A sketch like this is relatively clean, and the drawing is very fast and completed in one step.

best quality, masterpiece, (realistic:1.2), 1 girl, brown hair, brown eyes,Front, detailed face, beautiful eyes,((look looking at viewer))

Finally, have you read this content yet? AI drawing practice (2): Make a picture book painting of the ocean, beach, boy and dog | Stable Diffusion becomes a productivity tool for designers