model/decoder.py

import torch
import torch.nn as nn
from torch.nn import init as init
import torch.nn.functional as F
from torch.nn.modules.batchnorm import _BatchNorm

class Decoder_Identity(nn.Module):
    def __init__(self):
        super(Decoder_Identity, self).__init__()

        self.conv_up_2 = nn.Sequential(
            nn.ConvTranspose2d(in_channels=64, out_channels=32, kernel_size=3, stride=2, padding=1, output_padding=1, dilation=1),
            nn.ReLU(),
            nn.Conv2d(in_channels=32, out_channels=32, kernel_size=3, padding=1, bias=True),
            nn.ReLU(),
            nn.Conv2d(in_channels=32, out_channels=32, kernel_size=3, padding=1, bias=True),
            nn.ReLU()
        )

        self.conv_up_1 = nn.Sequential(
            nn.ConvTranspose2d(in_channels=32, out_channels=16, kernel_size=3, stride=2, padding=1, output_padding=1, dilation=1),
            nn.ReLU(),
            nn.Conv2d(in_channels=16, out_channels=16, kernel_size=3, padding=1, bias=True),
            nn.ReLU(),
            nn.Conv2d(in_channels=16, out_channels=16, kernel_size=3, padding=1, bias=True),
            nn.ReLU()
        )

        self.conv_last = nn.Sequential(
            nn.Conv2d(in_channels=16, out_channels=3, kernel_size=1, bias=True),
            nn.ReLU()
        )

    def forward(self, feat):
        featmap_2 = self.conv_up_2(feat)
        featmap_1 = self.conv_up_1(featmap_2)
        out = self.conv_last(featmap_1)

        return out


class Decoder_SR(nn.Module):
    def __init__(self, scale=4):
        super(Decoder_SR, self).__init__()

        self.scale = scale

        self.conv_up_2 = nn.Sequential(
            nn.ConvTranspose2d(in_channels=64, out_channels=32, kernel_size=3, stride=2, padding=1, output_padding=1, dilation=1),
            nn.ReLU(),
            nn.Conv2d(in_channels=32, out_channels=32, kernel_size=3, padding=1, bias=True),
            nn.ReLU(),
            nn.Conv2d(in_channels=32, out_channels=32, kernel_size=3, padding=1, bias=True),
            nn.ReLU()
        )

        self.conv_up_1 = nn.Sequential(
            nn.ConvTranspose2d(in_channels=32, out_channels=16, kernel_size=3, stride=2, padding=1, output_padding=1, dilation=1),
            nn.ReLU(),
            nn.Conv2d(in_channels=16, out_channels=16, kernel_size=3, padding=1, bias=True),
            nn.ReLU(),
            nn.Conv2d(in_channels=16, out_channels=16, kernel_size=3, padding=1, bias=True),
            nn.ReLU()
        )

        self.lrelu = nn.LeakyReLU(negative_slope=0.2, inplace=True)

        # upsampling
        self.upsample_1 = nn.Conv2d(in_channels=16, out_channels=16, kernel_size=3, padding=1, bias=True)
        self.upsample_2 = nn.Conv2d(in_channels=16, out_channels=16, kernel_size=3, padding=1, bias=True)

        self.HR_conv = nn.Conv2d(in_channels=16, out_channels=16, kernel_size=3, padding=1, bias=True)
        self.conv_last = nn.Conv2d(in_channels=16, out_channels=3, kernel_size=3, padding=1, bias=True)

    def forward(self, feat):
        featmap_2 = self.conv_up_2(feat)
        featmap_1 = self.conv_up_1(featmap_2)

        if self.scale == 4:
            featmap = self.lrelu(self.upsample_1(F.interpolate(featmap_1, scale_factor=2, mode='nearest')))
            featmap = self.lrelu(self.upsample_2(F.interpolate(featmap, scale_factor=2, mode='nearest')))
        elif self.scale == 2:
            featmap = self.lrelu(self.upsample_1(F.interpolate(featmap_1, scale_factor=2, mode='nearest')))


        out = self.conv_last(self.lrelu(self.HR_conv(featmap)))

        return out


def default_init_weights(module_list, scale=1, bias_fill=0, **kwargs):
    """Initialize network weights.

    Args:
        module_list (list[nn.Module] | nn.Module): Modules to be initialized.
        scale (float): Scale initialized weights, especially for residual
            blocks. Default: 1.
        bias_fill (float): The value to fill bias. Default: 0
        kwargs (dict): Other arguments for initialization function.
    """
    if not isinstance(module_list, list):
        module_list = [module_list]
    for module in module_list:
        for m in module.modules():
            if isinstance(m, nn.Conv2d):
                init.kaiming_normal_(m.weight, **kwargs)
                m.weight.data *= scale
                if m.bias is not None:
                    m.bias.data.fill_(bias_fill)
            elif isinstance(m, nn.Linear):
                init.kaiming_normal_(m.weight, **kwargs)
                m.weight.data *= scale
                if m.bias is not None:
                    m.bias.data.fill_(bias_fill)
            elif isinstance(m, _BatchNorm):
                init.constant_(m.weight, 1)
                if m.bias is not None:
                    m.bias.data.fill_(bias_fill)


def make_layer(basic_block, num_basic_block, **kwarg):
    """Make layers by stacking the same blocks.

    Args:
        basic_block (nn.module): nn.module class for basic block.
        num_basic_block (int): number of blocks.

    Returns:
        nn.Sequential: Stacked blocks in nn.Sequential.
    """
    layers = []
    for _ in range(num_basic_block):
        layers.append(basic_block(**kwarg))
    return nn.Sequential(*layers)


class ResidualDenseBlock(nn.Module):
    """Residual Dense Block.

    Used in RRDB block in ESRGAN.

    Args:
        num_feat (int): Channel number of intermediate features.
        num_grow_ch (int): Channels for each growth.
    """

    def __init__(self, num_feat=64, num_grow_ch=32):
        super(ResidualDenseBlock, self).__init__()
        self.conv1 = nn.Conv2d(num_feat, num_grow_ch, 3, 1, 1)
        self.conv2 = nn.Conv2d(num_feat + num_grow_ch, num_grow_ch, 3, 1, 1)
        self.conv3 = nn.Conv2d(num_feat + 2 * num_grow_ch, num_grow_ch, 3, 1, 1)
        self.conv4 = nn.Conv2d(num_feat + 3 * num_grow_ch, num_grow_ch, 3, 1, 1)
        self.conv5 = nn.Conv2d(num_feat + 4 * num_grow_ch, num_feat, 3, 1, 1)

        self.lrelu = nn.LeakyReLU(negative_slope=0.2, inplace=True)

        # initialization
        default_init_weights([self.conv1, self.conv2, self.conv3, self.conv4, self.conv5], 0.1)

    def forward(self, x):
        x1 = self.lrelu(self.conv1(x))
        x2 = self.lrelu(self.conv2(torch.cat((x, x1), 1)))
        x3 = self.lrelu(self.conv3(torch.cat((x, x1, x2), 1)))
        x4 = self.lrelu(self.conv4(torch.cat((x, x1, x2, x3), 1)))
        x5 = self.conv5(torch.cat((x, x1, x2, x3, x4), 1))
        # Emperically, we use 0.2 to scale the residual for better performance
        return x5 * 0.2 + x


class RRDB(nn.Module):
    """Residual in Residual Dense Block.

    Used in RRDB-Net in ESRGAN.

    Args:
        num_feat (int): Channel number of intermediate features.
        num_grow_ch (int): Channels for each growth.
    """

    def __init__(self, num_feat, num_grow_ch=32):
        super(RRDB, self).__init__()
        self.rdb1 = ResidualDenseBlock(num_feat, num_grow_ch)
        self.rdb2 = ResidualDenseBlock(num_feat, num_grow_ch)
        self.rdb3 = ResidualDenseBlock(num_feat, num_grow_ch)

    def forward(self, x):
        out = self.rdb1(x)
        out = self.rdb2(out)
        out = self.rdb3(out)
        # Emperically, we use 0.2 to scale the residual for better performance
        return out * 0.2 + x


class Decoder_Id_RRDB(nn.Module):
    def __init__(self, num_in_ch, num_out_ch=3, scale=4, num_feat=64, num_block=10, num_grow_ch=32):
        super(Decoder_Id_RRDB, self).__init__()

        self.conv_first = nn.Conv2d(num_in_ch, num_feat, 3, 1, 1)
        self.body = make_layer(RRDB, num_block, num_feat=num_feat, num_grow_ch=num_grow_ch)
        self.conv_body = nn.Conv2d(num_feat, num_feat, 3, 1, 1)
        
        self.conv_hr = nn.Conv2d(num_feat, num_feat, 3, 1, 1)
        self.conv_last = nn.Conv2d(num_feat, num_out_ch, 3, 1, 1)

        self.lrelu = nn.LeakyReLU(negative_slope=0.2, inplace=True)
    
    def forward(self, x):

        feat = self.conv_first(x)
        body_feat = self.conv_body(self.body(feat))
        feat = feat + body_feat
        
        out = self.conv_last(self.lrelu(self.conv_hr(feat)))
        return out


class Decoder_SR_RRDB(nn.Module):
    def __init__(self, num_in_ch, num_out_ch=3, scale=4, num_feat=64, num_block=23, num_grow_ch=32):
        super(Decoder_SR_RRDB, self).__init__()
        self.conv_first = nn.Conv2d(num_in_ch, num_feat, 3, 1, 1)
        self.body = make_layer(RRDB, num_block, num_feat=num_feat, num_grow_ch=num_grow_ch)
        self.conv_body = nn.Conv2d(num_feat, num_feat, 3, 1, 1)
        # upsample
        self.conv_up1 = nn.Conv2d(num_feat, num_feat, 3, 1, 1)
        self.conv_up2 = nn.Conv2d(num_feat, num_feat, 3, 1, 1)
        self.conv_hr = nn.Conv2d(num_feat, num_feat, 3, 1, 1)
        self.conv_last = nn.Conv2d(num_feat, num_out_ch, 3, 1, 1)

        self.lrelu = nn.LeakyReLU(negative_slope=0.2, inplace=True)

    def forward(self, x):

        feat = self.conv_first(x)
        body_feat = self.conv_body(self.body(feat))
        feat = feat + body_feat
        # upsample
        feat = self.lrelu(self.conv_up1(F.interpolate(feat, scale_factor=2, mode='nearest')))
        feat = self.lrelu(self.conv_up2(F.interpolate(feat, scale_factor=2, mode='nearest')))
        out = self.conv_last(self.lrelu(self.conv_hr(feat)))
        return out