93 lines
3.1 KiB
Python
93 lines
3.1 KiB
Python
import unfake
|
|
import torch
|
|
import numpy as np
|
|
from PIL import Image
|
|
import nest_asyncio
|
|
nest_asyncio.apply()
|
|
|
|
def _tensor_to_pil(image: torch.Tensor) -> Image.Image:
|
|
if len(image.shape) > 3:
|
|
image = image[0]
|
|
input_tensor = image.cpu()
|
|
image_np = (input_tensor.numpy() * 255).astype(np.uint8)
|
|
if image_np.shape[-1] == 4:
|
|
mode = "RGBA"
|
|
elif image_np.shape[-1] == 3:
|
|
mode = "RGB"
|
|
else:
|
|
mode = None
|
|
return Image.fromarray(image_np, mode=mode)
|
|
|
|
def _pil_to_tensor(img: Image.Image) -> torch.Tensor:
|
|
np_image = np.array(img).astype(np.float32) / 255.0
|
|
return torch.from_numpy(np_image).unsqueeze(0)
|
|
|
|
class CustomUnfake():
|
|
|
|
def __init__(self):
|
|
pass
|
|
|
|
@classmethod
|
|
def INPUT_TYPES(cls):
|
|
|
|
return {
|
|
"required": {
|
|
"image": ("IMAGE",),
|
|
"max_colors": ("INT", {
|
|
"default": 0,
|
|
"min": 0,
|
|
"max": 256,
|
|
"step": 1,
|
|
"display": "number"
|
|
}),
|
|
"detect_method": (["auto", "runs", "edge"], {"default": "auto"}),
|
|
"downscale_method": (["dominant", "median", "mode", "mean", "content-adaptive"], {"default": "dominant"}),
|
|
"cleanup_morph": ("BOOLEAN", {"default": True}),
|
|
"cleanup_jaggies": ("BOOLEAN", {"default": False}),
|
|
"transparent_background": ("BOOLEAN", {"default": False}),
|
|
},
|
|
}
|
|
|
|
RETURN_TYPES = ("IMAGE","STRING")
|
|
RETURN_NAMES = ("align_img", "align_info")
|
|
FUNCTION = "unfaked_image"
|
|
CATEGORY = "BillBum/Image"
|
|
|
|
def unfaked_image(self, image, max_colors, detect_method, downscale_method, cleanup_morph, cleanup_jaggies, transparent_background):
|
|
|
|
pil_img = _tensor_to_pil(image)
|
|
|
|
if max_colors == 0:
|
|
result = unfake.process_image_sync(
|
|
pil_img,
|
|
detect_method=detect_method,
|
|
downscale_method=downscale_method,
|
|
cleanup={"morph": cleanup_morph, "jaggy": cleanup_jaggies},
|
|
snap_grid=True,
|
|
transparent_background=transparent_background
|
|
)
|
|
else:
|
|
result = unfake.process_image_sync(
|
|
pil_img,
|
|
max_colors=max_colors,
|
|
detect_method=detect_method,
|
|
downscale_method=downscale_method,
|
|
cleanup={"morph": cleanup_morph, "jaggy": cleanup_jaggies},
|
|
snap_grid=True,
|
|
transparent_background=transparent_background
|
|
)
|
|
manifest = result["manifest"]
|
|
width, height = manifest.final_size
|
|
final_size = f"{width}x{height}"
|
|
final_colors = manifest.processing_steps["color_quantization"]["final_colors"]
|
|
out_info = f"{final_size}px_c{final_colors}"
|
|
result_tensor = _pil_to_tensor(result["image"])
|
|
return (result_tensor, out_info)
|
|
|
|
## Meneger Mapping
|
|
NODE_CLASS_MAPPINGS = {
|
|
"CustomUnfake": CustomUnfake,
|
|
}
|
|
|
|
NODE_DISPLAY_NAME_MAPPINGS = {
|
|
} |