Video tutorial - 视频教程
Master Efficient workflow in a Snap - 秒会高效工作流
Resource:
Efficient node github or Install it by using ComfyUI Manager
ControlNet workflow - ControlNet工作流
ControlNet is a powerful tool that enhances the capabilities of diffusion models like Stable Diffusion in ComfyUI. It allows you to guide the image generation process by providing extra information, such as edges, depth maps, human poses, and more. This additional conditioning enables you to steer the generated imagery more precisely according to both textual and visual inputs.
ControlNet works by introducing a clone of the model's encoding layers for training, using "zero convolutions." These specially designed convolutional layers start with zero weights and carefully integrate the new spatial conditions without disrupting the model's original proficiency.
Resource:
ComfyUI ControlNet Example
ControlNet-v1-1
ControlNet-v1-1_fp16_safetensors
sd_control_collection
