91 lines
3.2 KiB
Markdown
91 lines
3.2 KiB
Markdown
|
|
# Documentation
|
|
- Class name: AV_ControlNetPreprocessor
|
|
- Category: Art Venture/Loaders
|
|
- Output node: False
|
|
- Repo Ref: https://github.com/comfyanonymous/ComfyUI
|
|
|
|
AV_ControlNetPreprocessor节点旨在为ControlNet框架内的进一步处理或分析准备图像。它允许对图像应用各种预处理技术，包括分辨率调整和特定预处理器的选择，以优化它们以满足ControlNet的要求。
|
|
|
|
# Input types
|
|
## Required
|
|
- image
|
|
- 需要预处理的输入图像。这是进行预处理操作的主要数据。
|
|
- Comfy dtype: IMAGE
|
|
- Python dtype: ImageType
|
|
- preprocessor
|
|
- 指定要应用于输入图像的预处理技术。它可以是几个预定义的预处理器之一或自定义预处理器，影响预处理结果。
|
|
- Comfy dtype: COMBO[STRING]
|
|
- Python dtype: str
|
|
- sd_version
|
|
- 指示要使用的Stable Diffusion的具体版本，影响预处理如何与模型的要求保持一致。
|
|
- Comfy dtype: COMBO[STRING]
|
|
- Python dtype: str
|
|
## Optional
|
|
- resolution
|
|
- 输出图像的所需分辨率，可以根据特定的质量或性能标准进行调整。
|
|
- Comfy dtype: INT
|
|
- Python dtype: int
|
|
- preprocessor_override
|
|
- 允许用替代预处理器覆盖最初选择的预处理器，为预处理选择提供灵活性。
|
|
- Comfy dtype: STRING
|
|
- Python dtype: str
|
|
|
|
# Output types
|
|
- IMAGE
|
|
- 预处理后的图像，准备在ControlNet框架内进行进一步处理或分析。
|
|
- Comfy dtype: IMAGE
|
|
- Python dtype: ImageType
|
|
- CNET_NAME
|
|
- 在预处理过程中使用或识别的ControlNet的名称，为后续处理步骤提供上下文。
|
|
- Comfy dtype: STRING
|
|
- Python dtype: str
|
|
|
|
|
|
## Usage tips
|
|
- Infra type: `CPU`
|
|
- Common nodes:
|
|
- [CR Multi-ControlNet Stack](../../ComfyUI_Comfyroll_CustomNodes/Nodes/CR Multi-ControlNet Stack.md)
|
|
|
|
|
|
|
|
## Source code
|
|
```python
|
|
class AV_ControlNetPreprocessor:
|
|
preprocessors = list(control_net_preprocessors.keys())
|
|
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"image": ("IMAGE",),
|
|
"preprocessor": (["None", "tile"] + s.preprocessors,),
|
|
"sd_version": (["sd15", "sd15++", "sdxl", "sdxl_t2i", "sdxl_lllite"],),
|
|
},
|
|
"optional": {
|
|
"resolution": ("INT", {"default": 512, "min": 64, "max": 2048, "step": 64}),
|
|
"preprocessor_override": ("STRING", {"default": "None"}),
|
|
},
|
|
}
|
|
|
|
RETURN_TYPES = ("IMAGE", "STRING")
|
|
RETURN_NAMES = ("IMAGE", "CNET_NAME")
|
|
FUNCTION = "detect_controlnet"
|
|
CATEGORY = "Art Venture/Loaders"
|
|
|
|
def detect_controlnet(self, image, preprocessor, sd_version, resolution=512, preprocessor_override="None"):
|
|
if preprocessor_override != "None":
|
|
if preprocessor_override not in control_net_preprocessors:
|
|
print(
|
|
f"Warning: Not found ControlNet preprocessor {preprocessor_override}. Use {preprocessor} instead."
|
|
)
|
|
else:
|
|
preprocessor = preprocessor_override
|
|
|
|
image = apply_preprocessor(image, preprocessor, resolution=resolution)
|
|
control_net_name = detect_controlnet(preprocessor, sd_version)
|
|
|
|
return (image, control_net_name)
|
|
|
|
```
|