IMPORTANT! IPAdapter was just updated, check this video if your updated IPAdapter is not working anymore: • IPAdapter v2 Released!...
Why pay for proprietary apps or scour the web for hours when you can generate infinite reference images for free?
In this tutorial, we'll see how we can generate any amount of refence images or brainstorming ideas with Stable Diffusion.
As always, we will see how this workflow can be applied in real production environments, with three example use cases: creating reference poses for photoshoots, creating reference images for moodboards, and creating brainstorming ideas for logo and graphic design.
If you like my contributions, you can buy me a coffee here: ko-fi.com/risunobushi
Ever wondered how to use Stable Diffusion in your traditional workflows in professional environments as a creative?
Stable Diffusion for Professional Creatives is a series of videos about how we can apply generative AI in production environments, with real life examples and guides.
Useful links (by category)
IMPORTANT: some models may have different filenames as the ones showed in the video. That's because I rename them after downloading in order to remember which model does its thing. Refer to the instruction in this description to properly set everything up.
Workflow:
- json file, import in comfyUI and install any missing nodes: pastebin.com/eKtQYbL3
Nodes (if you haven't installed them via importing the provided workflow):
- ComfyUI IPAdapter Plus GitHub: github.com/cubiq/ComfyUI_IPAd...
Models:
- Aetherverse XL (SDXL model used in this video): civitai.com/models/308337
- Dreamshaper 1.5 (if you want to use a 1.5 model, this is a good all-around): civitai.com/models/4384/dream...
IPAdapter (Read the instructions. These models may have slightly different denominations than the ones used in this video, as I rename my models to remember which are which):
- IPAdapter Plus XL model. Place it into your "\models\ipadapter" folder and use it in your Load IPAdapter Model node: huggingface.co/h94/IP-Adapter...
- IPAdapter Plus 1.5 model. Place it into your "\models\ipadapter" folder and use it in your Load IPAdapter Model node: huggingface.co/h94/IP-Adapter...
- CLIPVision ViT-H model (works with IPAdapter *PLUS*, both 1.5 and XL, only. For IPAdapter *STANDARD*, not used in this video and kind of deprecated, you need a ViT-G model). Place it into "\models\clip_vision" and use it in your Load CLIPVision Model node: huggingface.co/h94/IP-Adapter...
Timestamps:
00:00 - Intro
00:50 - Node Overview
02:05 - Setting up the node workflow in comfyUI
05:08 - Generating reference images for moodboards
11:06 - Generating poses for preproduction references
14:31 - Generating brainstorming ideas for logo / graphic design
17:23 - Outro
#stablediffusion #stablediffusiontutorial #ai #generativeai #generativeart #comfyui #comfyuitutorial #risunobushi_ai #moodboards #reference #sdxl #sd #risunobushi #andreabaioni