172 lines
6.5 KiB
Python
172 lines
6.5 KiB
Python
# -*- coding: utf-8 -*-
|
|
# Copyright (c) Alibaba, Inc. and its affiliates.
|
|
from PIL import Image
|
|
import numpy as np
|
|
import math
|
|
import torch
|
|
import torchvision.transforms as T
|
|
|
|
from scepter.modules.utils.config import Config
|
|
from scepter.modules.utils.distribute import we
|
|
from scepter.modules.annotator.registry import ANNOTATORS
|
|
from .constant import WORKFLOW_CONFIG
|
|
|
|
|
|
class ACEPlusProcessorNode:
|
|
def __init__(self,
|
|
max_aspect_ratio=4,
|
|
d=16,
|
|
processor=WORKFLOW_CONFIG.ace_plus_processor_config):
|
|
self.max_aspect_ratio = max_aspect_ratio
|
|
self.processor_cfg = processor
|
|
self.task_list = {}
|
|
self.d = d
|
|
self.max_seq_len = processor.DEFAULT_PARAS.MAX_SEQ_LENGTH
|
|
self.transforms = T.Compose([
|
|
T.ToTensor(),
|
|
T.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])
|
|
])
|
|
for task in self.processor_cfg.PROCESSORS:
|
|
self.task_list[task.TYPE] = task
|
|
|
|
CATEGORY = '🪄 ComfyUI-Scepter'
|
|
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
'required': {
|
|
'ref_image': ('IMAGE',),
|
|
'task_type': (list(s().task_list.keys()),),
|
|
'repainting_scale': ('FLOAT', {
|
|
'default': 1,
|
|
'min': 0,
|
|
'max': 1,
|
|
'step': 0.01
|
|
}),
|
|
},
|
|
'optional': {
|
|
'edit_mask': ('MASK',),
|
|
'edit_image': ('IMAGE',)
|
|
}
|
|
}
|
|
|
|
OUTPUT_NODE = True
|
|
RETURN_TYPES = ('IMAGE', 'MASK')
|
|
RETURN_NAMES = ('IMAGE', 'MASK')
|
|
FUNCTION = 'execute'
|
|
|
|
def execute(self,
|
|
ref_image,
|
|
task_type,
|
|
edit_mask=None,
|
|
edit_image=None,
|
|
repainting_scale=1):
|
|
if task_type != 'image_processor':
|
|
edit_image = self.edit_preprocess(self.task_list[task_type], we.device_id,
|
|
ref_image, edit_mask)
|
|
return self.preprocess(ref_image, edit_image,
|
|
edit_mask, repainting_scale)
|
|
|
|
def edit_preprocess(self, processor, device, edit_image, edit_mask):
|
|
if edit_image is None or processor is None:
|
|
return edit_image
|
|
processor = Config(cfg_dict=processor, load=False)
|
|
processor = ANNOTATORS.build(processor).to(device)
|
|
edit_image = self.trans_tensor_pil(edit_image)
|
|
new_edit_image = processor(np.asarray(edit_image))
|
|
|
|
del processor
|
|
new_edit_image = Image.fromarray(new_edit_image)
|
|
to_pil = T.ToPILImage()
|
|
edit_mask = to_pil(edit_mask)
|
|
|
|
if new_edit_image.size != edit_image.size:
|
|
edit_image = T.Resize((edit_image.size[1], edit_image.size[0]),
|
|
interpolation=T.InterpolationMode.BILINEAR,
|
|
antialias=True)(new_edit_image)
|
|
image = Image.composite(new_edit_image, edit_image, edit_mask)
|
|
return self.trans_pil_tensor(image)
|
|
|
|
def trans_tensor_pil(self, tensor_image):
|
|
image = tensor_image.squeeze(0).permute(2, 0, 1)
|
|
to_pil = T.ToPILImage()
|
|
return to_pil(image)
|
|
|
|
def trans_pil_tensor(self, pil_image):
|
|
transform = T.Compose([
|
|
T.ToTensor()
|
|
])
|
|
tensor_image = transform(pil_image)
|
|
tensor_image = tensor_image.unsqueeze(0)
|
|
tensor_image = tensor_image.permute(0, 2, 3, 1)
|
|
|
|
return tensor_image
|
|
|
|
def image_check(self, image):
|
|
if image is None:
|
|
return image
|
|
W, H = image.size
|
|
if H / W > self.max_aspect_ratio:
|
|
image = T.CenterCrop([int(self.max_aspect_ratio * W), W])(image)
|
|
elif W / H > self.max_aspect_ratio:
|
|
image = T.CenterCrop([H, int(self.max_aspect_ratio * H)])(image)
|
|
return self.transforms(image)
|
|
|
|
def denormalize(self, t):
|
|
mean = torch.tensor([0.5, 0.5, 0.5]).view(-1, 1, 1)
|
|
std = torch.tensor([0.5, 0.5, 0.5]).view(-1, 1, 1)
|
|
return t * std + mean
|
|
|
|
def preprocess(self,
|
|
reference_image=None,
|
|
edit_image=None,
|
|
edit_mask=None,
|
|
repainting_scale=1.0):
|
|
reference_image = self.trans_tensor_pil(reference_image) \
|
|
if reference_image is not None else None
|
|
width, height = reference_image.size
|
|
edit_image = edit_image.squeeze(0).permute(2, 0, 1) \
|
|
if edit_image is not None else None
|
|
to_pil = T.ToPILImage()
|
|
edit_mask = to_pil(edit_mask) if edit_mask is not None else None
|
|
|
|
reference_image = self.image_check(reference_image)
|
|
if edit_image is None:
|
|
edit_image = torch.zeros([3, height, width])
|
|
edit_mask = torch.ones([1, height, width])
|
|
else:
|
|
edit_mask = np.asarray(edit_mask)
|
|
edit_mask = np.where(edit_mask > 128, 1, 0)
|
|
edit_mask = edit_mask.astype(
|
|
np.float32) if np.any(edit_mask) else np.ones_like(edit_mask).astype(
|
|
np.float32)
|
|
edit_mask = torch.tensor(edit_mask).unsqueeze(0)
|
|
|
|
edit_image = edit_image * (1 - edit_mask * repainting_scale)
|
|
|
|
assert edit_mask is not None
|
|
if reference_image is not None:
|
|
_, H, W = reference_image.shape
|
|
_, eH, eW = edit_image.shape
|
|
scale = eH / H
|
|
tH, tW = eH, int(W * scale)
|
|
reference_image = T.Resize((tH, tW),
|
|
interpolation=T.InterpolationMode.BILINEAR,
|
|
antialias=True)(reference_image)
|
|
if repainting_scale == 1:
|
|
reference_image = self.denormalize(reference_image)
|
|
edit_image = torch.cat([reference_image, edit_image], dim=-1)
|
|
edit_mask = torch.cat([torch.zeros([1, reference_image.shape[1],
|
|
reference_image.shape[2]]), edit_mask], dim=-1)
|
|
|
|
H, W = edit_image.shape[-2:]
|
|
scale = min(1.0, math.sqrt(self.max_seq_len * 2 / ((H / self.d) * (W / self.d))))
|
|
rH = int(H * scale) // self.d * self.d
|
|
rW = int(W * scale) // self.d * self.d
|
|
|
|
edit_image = T.Resize((rH, rW), interpolation=T.InterpolationMode.BILINEAR, antialias=True)(edit_image)
|
|
edit_mask = T.Resize((rH, rW), interpolation=T.InterpolationMode.NEAREST_EXACT, antialias=True)(edit_mask)
|
|
edit_image = edit_image.unsqueeze(0).permute(0, 2, 3, 1)
|
|
|
|
return edit_image, edit_mask
|