main reference
原作者His blog post: AI-generated scannable image - new ControlNet model display
Lighting generation model C station description: https://civitai.com/models/80536/lighting-based-picture-control-controlnet
Extended reading: https://aigc .ioclab.com/sd-showcase/light_controlnet.html
Effect display of others: https://www.pinterest.co.uk/technonatorr/qr-codes-produced-by-ai/
Controlnet model download address: https://huggingface .co/lllyasviel/ControlNet-v1-1/tree/main
Hands-on tutorial:
Reprinted video: https://www.bilibili.com/video/BV1yN411S7Tx
Note : The effect difference in the test comes from SD基础模型andVAE
DIY steps
You need to install the stable-diffusion-webui and controlnet plug-ins first, as well as the basic model. The figure
below shows the actual generation process, and there are failures in the middle of the generation. Need 仔细检查参数, see the screenshot for details
1. Generate a QR code online (you can use Baidu yourself)
- https://cli.im/url
本博文Use: https://qr.io/
2.1 Controlnet related model download
See controlnet
ui安装: https://github.com/Mikubill/sd-webui-controlnet
Lighting control model: control_v1p_sd15_brightness.safetensors
control_11_tile: https://huggingface.co/lllyasviel/ControlNet-v1-1/blob/main/control_v11f1e_sd15_tile.pth
2.2 vae download ( 可跳过, if you want to be exactly the same as this article)
Link: https://pan.baidu.com/s/1b9nhcTJs60-Z4CsCPucchw?pwd=0619
Name: kawaii2D_kawaii2DPrunedFp16.vae
You can also use other animation vae, the effect is slightly different
The default anything-v4.5-pruned.vae.pt the color will be worse
3 parameter setting
3.1 General parameters
base Model: anything-v4.5-pruned (you can choose by yourself)
Important: Sampler:DPM++ 2M Karras
a girl,best quality, ultra-detailed, extremely detailed,extremely delicate and beautiful, floating, high resolution, (masterpiece, best quality, very detailed, Ultra HD: 1.2)
Negative prompt: nsfw, (worst quality, bad quality:1.6), lowres, signature, watermark, username, easynegative
Steps: 25, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 3447199121, Size: 512x512, Model hash: 6e430eb514, Model: anything-v4.5-pruned, Clip skip: 2, ENSD: 31337, ControlNet 0: “preprocessor: inpaint_global_harmonious, model: control_v1p_sd15_brightness [5f6aa6ed], weight: 0.35, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (-1, 64, 64)”, ControlNet 1: “preprocessor: inpaint_global_harmonious, model: control_v11f1e_sd15_tile [a371b31b], weight: 0.65, starting/ending: (0.35, 0.75), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (-1, 64, 64)”
Upload the generated image to obtain the generated information
3.2 Setting of controlnet -brightness model
model: control_v1p_sd15_brightness [5f6aa6ed], weight: 0.35, starting/ending: (0, 1)
3.3 Settings of controlnet-tile model
ControlNet 1:
preprocessor: inpaint_global_harmonious,
model: control_v11f1e_sd15_tile [a371b31b],
weight: 0.65, starting/ending: (0.35, 0.75),
4. Other results after inference
