One of the biggest problems when it comes to Generative AI, is the fact that your are captive to the mercy of the AI Model.
You might create awesome prompts, use LORA to embed a sepcific style of character, but eventually, what comes out is based on the random seed the AI used for the image generation.
ControlNet solves just that! At first, it might seem like a very complex and steep thing to learn, but in this video, I will simplify the concept behind ControlNet and will show you how easy it is to get control over you text to image generations!
You can create anything your imagination can think of with very little technicalities.
You want to create a beautiful image? no problem.
You want to generate a unique sign for your friends birthday? no problem.
You want to add life to old photos? no problem!
Hop on in, and see how easy it is to rule and architect your AI Model!
P.S. Don't forget to like, subscribe, and share the good vibes ❤️
🔗 The-AI-Art Website: www.the-ai-art.com/
🔗 The AI-Art Discord: / discord
🔗 ComfyUI: github.com/comfyanonymous/Com...
🔗 Official XL Models: huggingface.co/stabilityai/co...
🔗 OpenPose XL Model: huggingface.co/thibaud/contro...
🔗 SD 1.5 Models: huggingface.co/lllyasviel/Con...
🔗 OpenPose generator: openposeai.com/
🔗 Pixlr editor: pixlr.com/express/
00:00 Intro
00:56 Getting started
01:20 How ControlNet works?
05:24 Finding reference images
06:20 Preprocessors? What? Where? How?
13:30 Applying ControlNet
17:35 Getting ControlNet Models
20:10 First use - Depth Map
22:10 Text Effects!
27:15 Pose pose pose!
30:20 Oh my god! they ki@%ed Canny ;)
31:10 Line Art
32:40 Using multiple ControlNet nodes
34:45 The magic of Recolor
38:52 Conclusion
#generativeart #ai #aiart #texttoimage #theaIart #the-ai-art #comfyui #stablediffusion #generativeai #generativeart #art #image #compare
#tutorial #training #controlnet
Негізгі бет Ғылым және технология ComfyUI: ControlNet simplified. The most powerful tool that gives you full control!
Пікірлер: 16