84 lines
2.9 KiB
Markdown
84 lines
2.9 KiB
Markdown
# Documentation
|
|
- Class name: Canvas_Tab
|
|
- Category: image
|
|
- Output node: False
|
|
- Repo Ref: https://github.com/Lerc/canvas_tab.git
|
|
|
|
Canvas_Tab 作为一个中介，处理和加工在一个单独标签页环境中的视觉内容数据。
|
|
|
|
# Input types
|
|
## Required
|
|
- unique_id
|
|
- 图像缓冲区实例的唯一标识符，对于在不同标签页或会话中管理和关联图像数据至关重要。
|
|
- Comfy dtype: str
|
|
- Python dtype: str
|
|
- mask
|
|
- 掩码参数对于定义图像中的兴趣区域至关重要，使得能够进行选择性处理和分析。
|
|
- Comfy dtype: str
|
|
- Python dtype: str
|
|
- canvas
|
|
- 画布参数代表源图像，它是所有图像操作和编辑操作的基础。
|
|
- Comfy dtype: str
|
|
- Python dtype: str
|
|
## Optional
|
|
- images
|
|
- 可选的图像集合，可用于比较或额外处理，增强图像缓冲区的多功能性。
|
|
- Comfy dtype: list
|
|
- Python dtype: list
|
|
|
|
# Output types
|
|
- rgb_image
|
|
- 处理后的RGB格式图像数据，准备进行进一步的分析或操作。
|
|
- Comfy dtype: tensor
|
|
- Python dtype: torch.Tensor
|
|
- mask_data
|
|
- 提取的掩码数据代表兴趣区域，对于目标图像分析和编辑至关重要。
|
|
- Comfy dtype: tensor
|
|
- Python dtype: torch.Tensor
|
|
|
|
# Usage tips
|
|
- Infra type: CPU
|
|
|
|
# Source code
|
|
```
|
|
class Canvas_Tab:
|
|
"""
|
|
A Image Buffer for handling an editor in another tab.
|
|
"""
|
|
|
|
def __init__(self):
|
|
pass
|
|
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {'required': {'mask': ('CANVAS',), 'canvas': ('CANVAS',)}, 'hidden': {'unique_id': 'UNIQUE_ID'}}
|
|
RETURN_TYPES = ('IMAGE', 'MASK')
|
|
FUNCTION = 'image_buffer'
|
|
CATEGORY = 'image'
|
|
|
|
def image_buffer(self, unique_id, mask, canvas, images=None):
|
|
print(f'-----hyn debug image_buffer : unique_id = {unique_id} canvas = {canvas} images = {images} masks = {mask}')
|
|
if canvas.startswith('http'):
|
|
from worker.components.utils import util
|
|
i = util.get_image_from_uri(canvas)
|
|
else:
|
|
image_path = folder_paths.get_annotated_filepath(canvas)
|
|
i = Image.open(image_path)
|
|
i = ImageOps.exif_transpose(i)
|
|
rgb_image = i.convert('RGB')
|
|
rgb_image = np.array(rgb_image).astype(np.float32) / 255.0
|
|
rgb_image = torch.from_numpy(rgb_image)[None,]
|
|
if mask.startswith('http'):
|
|
from worker.components.utils import util
|
|
i = util.get_image_from_uri(mask)
|
|
else:
|
|
mask_path = folder_paths.get_annotated_filepath(mask)
|
|
i = Image.open(mask_path)
|
|
i = ImageOps.exif_transpose(i)
|
|
if 'A' in i.getbands():
|
|
mask_data = np.array(i.getchannel('A')).astype(np.float32) / 255.0
|
|
mask_data = torch.from_numpy(mask_data)
|
|
else:
|
|
mask_data = torch.zeros((64, 64), dtype=torch.float32, device='cpu')
|
|
return (rgb_image, mask_data)
|
|
``` |