网络结构
该网络将输入图像通过Guided filter分解为Detail layer和Base layer
Detail layer:restoring high frequency details(SR)
Base layer:recovering local contrast(ITM)
Base layer即为低频:低频做逆色调映射
Detail layer高频:高频做超分辨率
网络的核心思想主要是这种图像分解做不同任务,最后通过融合,得到最终的效果。网络细节部分主要是一些模块的叠加,主要为如下几大模块:
(1)ResBlock
```python
class BaseBlock(nn.Module):
def __init__(self, channels):
super(BaseBlock, self).__init__()
self.conv1 = convBlock_c(channels=channels)
self.conv2 = convBlock_c(channels=channels)
self.relu = reluBlock()
def forward(self, x):
out = self.relu(x)
out = self.conv1(out)
out = self.relu(out)
out = self.conv2(out)
return out
class ResBlock(nn.Module):
def __init__(self, channels