【论文笔记】图像修复MPRNet:Multi-Stage Progressive Image Restoration 含代码解析

阿里云国内75折 回扣 微信号：monov8
阿里云国际，腾讯云国际，低至75折。AWS 93折 免费开户实名账号 代冲值 优惠多多 微信号：monov8 飞机：@monov6

目录

一、介绍

二、使用方法

1.推理

2.训练

三、MPRNet结构

1.整体结构

2.CAB(Channel Attention Block)

3.Stage1 Encoder

4.Stage2 Encoder

5.Decoder

6.SAM(Supervised Attention Module)

7.ORSNet(Original Resolution Subnetwork)

四、损失函数

1.CharbonnierLoss

2.EdgeLoss

一、介绍

        论文地址https://arxiv.org/pdf/2102.02808.pdf

        代码地址http://github.com/swz30/MPRNet

        恢复图像任务需要在空间细节和高级上下文特征之间取得复杂的平衡。于是作者设计了一个多阶段的模型模型首先使用编解码器架构来学习上下文的特征然后将它们与保留局部信息的高分辨率分支结合起来。

        打个比方我要修复一张蛇的图片编解码器负责提取高级上下文特征告诉模型要在蛇身上“画”鳞片而不是羽毛或其他东西然后高分辨率分支负责细化鳞片的图案。

        MPRNet细节很多但最主要的创新还是“多阶段”模型共有三个阶段前两个阶段是编解码器子网络用来学习较大感受野的上下文特征最后一个阶段是高分辨率分支用于在最终的输出图像中构建所需的纹理。作者给出了Deblurring、Denoising、Deraining三个任务的项目三个项目的backbone是一样的只是参数规模有所不同(Deblurring>Denoising>Deraining)下面我们以最大的Deblurring为例进行介绍。

二、使用方法

        MPRNet项目分为Deblurring、Denoising和Deraining  三个子项目。作者没有用稀奇古怪的库也没用高级的编程技巧非常适合拿来研究学习使用方法也很简单几句话技能说完。

1.推理

        1下载预训练模型预训练模型分别存在三个子项目的pretrained_models文件夹下载地址在每个pretrained_models文件夹的 README.md中需要科学上网我放在了网盘里

        链接https://pan.baidu.com/s/1sxfidMvlU_pIeO5zD1tKZg 提取码faye 

        2准备测试图片将退化图片放在目录samples/input/中

        3执行demo.py

# 执行Deblurring
python demo.py --task  Deblurring

# 执行Denoising
python demo.py --task  Denoising

# 执行Deraining
python demo.py --task  Deraining

        4结果放在目录samples/output/中。

2.训练

        1根据Dataset文件夹内的README.md文件中的地址下载数据集。

        2查看training.yml是否需要修改主要是最后的数据集地址。

        3执行训练

python train.py

三、MPRNet结构

        我将按照官方代码实现来介绍模型结构一些重要模块的划分可能跟论文有区别但是整体结构是一样的。

1.整体结构

        MPRNet官方给出的结构图如下

图1

        这个图总体概括了MPRNet的结构但是很多细节没有表现出来通过阅读代码我给出更加详细的模型结构介绍。下面的图中输入统一512x512我们以Deblurring为例并且batch_size=1。

        整体结构图如下

 图2

        图中的三个Input都是原图整个模型三个阶段整体流程如下

        1.1 输入图片采用multi-patch方式分成四份分成左上、右上、左下、右下

        1.2 每个patch经过一个3x3的卷积扩充维度为的是后面能提取更丰富的特征信息

        1.3 经过CAB(Channel Attention Block)利用注意力机制提取每个维度上的特征

        1.4 Encoder编码三种尺度的图像特征提取多尺度上下文特征同时也是提取更深层的语义特征

        1.5 合并深特征将四个batch的同尺度特征合并成左右两个尺度送入Decoder

        1.6 Decoder提取合并后的每个尺度的特征

        1.7 输入图片采用multi-patch方式分成两份分成左、右

        1.8 将左右两个batch分别与Stage1 Decoder输出的大尺度特征图送入SAMSupervised Attention ModuleSAM在训练的时候可以利用GT为当前阶段的恢复过程提供有用的控制信号

        1.9 SAM的输出分成两部分一部分是第二次输入的原图特征它将继续下面的流程一部分用于训练时的Stage1输出可以利用GT更快更好的让模型收敛。

        2.0 经过Stage2的卷积扩充通道和CAB操作将Stage1中的Decoder前后的特征送入Stage2的Encoder。

        2.2 经过和Stage1相似的Decoder也产生两个部分的输出一部分继续Stage3一部分输出与GT算损失

        3.1 Stage3的原图输入不在切分目的是利用完整的上下文信息恢复图片细节。

        3.2 将原图经过卷积做升维处理

        3.3 将Stage2中的Decoder前后的特征送入Stage3的ORSNet(Original Resolution Subnetwork)ORSNet不使用任何降采样操作并生成空间丰富的高分辨率特征。

        3.4 最后经过一个卷积将维度降为3输出。

代码实现

#位置MPRNet.py
class MPRNet(nn.Module):
    def __init__(self, in_c=3, out_c=3, n_feat=96, scale_unetfeats=48, scale_orsnetfeats=32, num_cab=8, kernel_size=3, reduction=4, bias=False):
        super(MPRNet, self).__init__()

        act=nn.PReLU()
        self.shallow_feat1 = nn.Sequential(conv(in_c, n_feat, kernel_size, bias=bias), CAB(n_feat,kernel_size, reduction, bias=bias, act=act))
        self.shallow_feat2 = nn.Sequential(conv(in_c, n_feat, kernel_size, bias=bias), CAB(n_feat,kernel_size, reduction, bias=bias, act=act))
        self.shallow_feat3 = nn.Sequential(conv(in_c, n_feat, kernel_size, bias=bias), CAB(n_feat,kernel_size, reduction, bias=bias, act=act))

        # Cross Stage Feature Fusion (CSFF)
        self.stage1_encoder = Encoder(n_feat, kernel_size, reduction, act, bias, scale_unetfeats, csff=False)
        self.stage1_decoder = Decoder(n_feat, kernel_size, reduction, act, bias, scale_unetfeats)

        self.stage2_encoder = Encoder(n_feat, kernel_size, reduction, act, bias, scale_unetfeats, csff=True)
        self.stage2_decoder = Decoder(n_feat, kernel_size, reduction, act, bias, scale_unetfeats)

        self.stage3_orsnet = ORSNet(n_feat, scale_orsnetfeats, kernel_size, reduction, act, bias, scale_unetfeats, num_cab)

        self.sam12 = SAM(n_feat, kernel_size=1, bias=bias)
        self.sam23 = SAM(n_feat, kernel_size=1, bias=bias)
        
        self.concat12  = conv(n_feat*2, n_feat, kernel_size, bias=bias)
        self.concat23  = conv(n_feat*2, n_feat+scale_orsnetfeats, kernel_size, bias=bias)
        self.tail     = conv(n_feat+scale_orsnetfeats, out_c, kernel_size, bias=bias)

    def forward(self, x3_img):
        # Original-resolution Image for Stage 3
        H = x3_img.size(2)
        W = x3_img.size(3)

        # Multi-Patch Hierarchy: Split Image into four non-overlapping patches

        # Two Patches for Stage 2
        x2top_img  = x3_img[:,:,0:int(H/2),:]
        x2bot_img  = x3_img[:,:,int(H/2):H,:]

        # Four Patches for Stage 1
        x1ltop_img = x2top_img[:,:,:,0:int(W/2)]
        x1rtop_img = x2top_img[:,:,:,int(W/2):W]
        x1lbot_img = x2bot_img[:,:,:,0:int(W/2)]
        x1rbot_img = x2bot_img[:,:,:,int(W/2):W]

        ##-------------------------------------------
        ##-------------- Stage 1---------------------
        ##-------------------------------------------
        ## Compute Shallow Features
        x1ltop = self.shallow_feat1(x1ltop_img)
        x1rtop = self.shallow_feat1(x1rtop_img)
        x1lbot = self.shallow_feat1(x1lbot_img)
        x1rbot = self.shallow_feat1(x1rbot_img)
        
        ## Process features of all 4 patches with Encoder of Stage 1
        feat1_ltop = self.stage1_encoder(x1ltop)
        feat1_rtop = self.stage1_encoder(x1rtop)
        feat1_lbot = self.stage1_encoder(x1lbot)
        feat1_rbot = self.stage1_encoder(x1rbot)
        
        ## Concat deep features
        feat1_top = [torch.cat((k,v), 3) for k,v in zip(feat1_ltop,feat1_rtop)]
        feat1_bot = [torch.cat((k,v), 3) for k,v in zip(feat1_lbot,feat1_rbot)]
        
        ## Pass features through Decoder of Stage 1
        res1_top = self.stage1_decoder(feat1_top)
        res1_bot = self.stage1_decoder(feat1_bot)

        ## Apply Supervised Attention Module (SAM)
        x2top_samfeats, stage1_img_top = self.sam12(res1_top[0], x2top_img)
        x2bot_samfeats, stage1_img_bot = self.sam12(res1_bot[0], x2bot_img)

        ## Output image at Stage 1
        stage1_img = torch.cat([stage1_img_top, stage1_img_bot],2) 
        ##-------------------------------------------
        ##-------------- Stage 2---------------------
        ##-------------------------------------------
        ## Compute Shallow Features
        x2top  = self.shallow_feat2(x2top_img)
        x2bot  = self.shallow_feat2(x2bot_img)

        ## Concatenate SAM features of Stage 1 with shallow features of Stage 2
        x2top_cat = self.concat12(torch.cat([x2top, x2top_samfeats], 1))
        x2bot_cat = self.concat12(torch.cat([x2bot, x2bot_samfeats], 1))

        ## Process features of both patches with Encoder of Stage 2
        feat2_top = self.stage2_encoder(x2top_cat, feat1_top, res1_top)
        feat2_bot = self.stage2_encoder(x2bot_cat, feat1_bot, res1_bot)

        ## Concat deep features
        feat2 = [torch.cat((k,v), 2) for k,v in zip(feat2_top,feat2_bot)]

        ## Pass features through Decoder of Stage 2
        res2 = self.stage2_decoder(feat2)

        ## Apply SAM
        x3_samfeats, stage2_img = self.sam23(res2[0], x3_img)


        ##-------------------------------------------
        ##-------------- Stage 3---------------------
        ##-------------------------------------------
        ## Compute Shallow Features
        x3     = self.shallow_feat3(x3_img)

        ## Concatenate SAM features of Stage 2 with shallow features of Stage 3
        x3_cat = self.concat23(torch.cat([x3, x3_samfeats], 1))
        
        x3_cat = self.stage3_orsnet(x3_cat, feat2, res2)

        stage3_img = self.tail(x3_cat)

        return [stage3_img+x3_img, stage2_img, stage1_img]

        图中还有一些模块细节没有表现出来下面我将详细介绍。

2.CAB(Channel Attention Block)

        顾名思义CAB就是利用注意力机制提取每个通道的特征输出输入特征图形状不变结构图如下

图3 

        可以看到经过了两个卷积和GAP之后得到了一个概率图就是那个残差边在经过两个卷积和Sigmoid之后与概率图相乘就实现了一个通道注意力机制。

        代码实现

# 位置MPRNet.py
## Channel Attention Layer
class CALayer(nn.Module):
    def __init__(self, channel, reduction=16, bias=False):
        super(CALayer, self).__init__()
        # global average pooling: feature --> point
        self.avg_pool = nn.AdaptiveAvgPool2d(1)
        # feature channel downscale and upscale --> channel weight
        self.conv_du = nn.Sequential(
                nn.Conv2d(channel, channel // reduction, 1, padding=0, bias=bias),
                nn.ReLU(inplace=True),
                nn.Conv2d(channel // reduction, channel, 1, padding=0, bias=bias),
                nn.Sigmoid()
        )

    def forward(self, x):
        y = self.avg_pool(x)
        y = self.conv_du(y)
        return x * y


##########################################################################
## Channel Attention Block (CAB)
class CAB(nn.Module):
    def __init__(self, n_feat, kernel_size, reduction, bias, act):
        super(CAB, self).__init__()
        modules_body = []
        modules_body.append(conv(n_feat, n_feat, kernel_size, bias=bias))
        modules_body.append(act)
        modules_body.append(conv(n_feat, n_feat, kernel_size, bias=bias))

        self.CA = CALayer(n_feat, reduction, bias=bias)
        self.body = nn.Sequential(*modules_body)

    def forward(self, x):
        res = self.body(x)
        res = self.CA(res)
        res += x
        return res

3.Stage1 Encoder

        Stage1和Stage1的Encoder有一些区别所以分开介绍。Stage1 Encoder有一个输入和三个不同尺度的输出为的是提取三个尺度的特征并为下面的尺度融合流程做准备其中有多个CAB结构可以更好的提取通道特征下采样通过粗暴的Downsample实现结构如下

         图4

        代码实现

# 位置MPRNet.py
class Encoder(nn.Module):
    def __init__(self, n_feat, kernel_size, reduction, act, bias, scale_unetfeats, csff):
        super(Encoder, self).__init__()

        self.encoder_level1 = [CAB(n_feat,                     kernel_size, reduction, bias=bias, act=act) for _ in range(2)]
        self.encoder_level2 = [CAB(n_feat+scale_unetfeats,     kernel_size, reduction, bias=bias, act=act) for _ in range(2)]
        self.encoder_level3 = [CAB(n_feat+(scale_unetfeats*2), kernel_size, reduction, bias=bias, act=act) for _ in range(2)]

        self.encoder_level1 = nn.Sequential(*self.encoder_level1)
        self.encoder_level2 = nn.Sequential(*self.encoder_level2)
        self.encoder_level3 = nn.Sequential(*self.encoder_level3)

        self.down12  = DownSample(n_feat, scale_unetfeats)
        self.down23  = DownSample(n_feat+scale_unetfeats, scale_unetfeats)

        # Cross Stage Feature Fusion (CSFF)
        if csff:
            self.csff_enc1 = nn.Conv2d(n_feat,                     n_feat,                     kernel_size=1, bias=bias)
            self.csff_enc2 = nn.Conv2d(n_feat+scale_unetfeats,     n_feat+scale_unetfeats,     kernel_size=1, bias=bias)
            self.csff_enc3 = nn.Conv2d(n_feat+(scale_unetfeats*2), n_feat+(scale_unetfeats*2), kernel_size=1, bias=bias)

            self.csff_dec1 = nn.Conv2d(n_feat,                     n_feat,                     kernel_size=1, bias=bias)
            self.csff_dec2 = nn.Conv2d(n_feat+scale_unetfeats,     n_feat+scale_unetfeats,     kernel_size=1, bias=bias)
            self.csff_dec3 = nn.Conv2d(n_feat+(scale_unetfeats*2), n_feat+(scale_unetfeats*2), kernel_size=1, bias=bias)

    def forward(self, x, encoder_outs=None, decoder_outs=None):
        enc1 = self.encoder_level1(x)
        if (encoder_outs is not None) and (decoder_outs is not None):
            enc1 = enc1 + self.csff_enc1(encoder_outs[0]) + self.csff_dec1(decoder_outs[0])

        x = self.down12(enc1)

        enc2 = self.encoder_level2(x)
        if (encoder_outs is not None) and (decoder_outs is not None):
            enc2 = enc2 + self.csff_enc2(encoder_outs[1]) + self.csff_dec2(decoder_outs[1])

        x = self.down23(enc2)

        enc3 = self.encoder_level3(x)
        if (encoder_outs is not None) and (decoder_outs is not None):
            enc3 = enc3 + self.csff_enc3(encoder_outs[2]) + self.csff_dec3(decoder_outs[2])
        
        return [enc1, enc2, enc3]

4.Stage2 Encoder

        Stage2 Encoder输入为三个分别为上一层的输出和Stage1中的Decoder前后的特征。主流程也就是左面竖着的那一列和Stage1 Encoder是一样的。增加的两个输入每个输入又分为三个尺度每个尺度经过一个卷积层然后相同尺度的特征图做特征融合输出结构如下

图5

5.Decoder

        两个阶段的Decoder结构是一样的所以放在一起说有三个不用尺度的输入通过CAB提取特征小尺度特征通过上采样变大通过卷积使通道变小小尺度的特征图shape最终变成跟大尺度一样通过残差边实现特征融合结构如下

 图6

        代码实现

# 位置MPRNet.py
class Decoder(nn.Module):
    def __init__(self, n_feat, kernel_size, reduction, act, bias, scale_unetfeats):
        super(Decoder, self).__init__()

        self.decoder_level1 = [CAB(n_feat,                     kernel_size, reduction, bias=bias, act=act) for _ in range(2)]
        self.decoder_level2 = [CAB(n_feat+scale_unetfeats,     kernel_size, reduction, bias=bias, act=act) for _ in range(2)]
        self.decoder_level3 = [CAB(n_feat+(scale_unetfeats*2), kernel_size, reduction, bias=bias, act=act) for _ in range(2)]

        self.decoder_level1 = nn.Sequential(*self.decoder_level1)
        self.decoder_level2 = nn.Sequential(*self.decoder_level2)
        self.decoder_level3 = nn.Sequential(*self.decoder_level3)

        self.skip_attn1 = CAB(n_feat,                 kernel_size, reduction, bias=bias, act=act)
        self.skip_attn2 = CAB(n_feat+scale_unetfeats, kernel_size, reduction, bias=bias, act=act)

        self.up21  = SkipUpSample(n_feat, scale_unetfeats)
        self.up32  = SkipUpSample(n_feat+scale_unetfeats, scale_unetfeats)

    def forward(self, outs):
        enc1, enc2, enc3 = outs
        dec3 = self.decoder_level3(enc3)

        x = self.up32(dec3, self.skip_attn2(enc2))
        dec2 = self.decoder_level2(x)

        x = self.up21(dec2, self.skip_attn1(enc1))
        dec1 = self.decoder_level1(x)

        return [dec1,dec2,dec3]

6.SAM(Supervised Attention Module)

        SAM出现在两个阶段间有两个输入将上一层特征和原图作为输入提升了特征提取的性能SAM作为有监督的注意模块使用注意力图强力筛选了跨阶段间的有用特征。有两个输出一个是经过了注意力机制的特征图为下面的流程提供特征一个是3通道的图片特征为了训练阶段输出结构如下

 图7

        代码位置

# 位置MPRNet.py
## Supervised Attention Module
class SAM(nn.Module):
    def __init__(self, n_feat, kernel_size, bias):
        super(SAM, self).__init__()
        self.conv1 = conv(n_feat, n_feat, kernel_size, bias=bias)
        self.conv2 = conv(n_feat, 3, kernel_size, bias=bias)
        self.conv3 = conv(3, n_feat, kernel_size, bias=bias)

    def forward(self, x, x_img):
        x1 = self.conv1(x)
        img = self.conv2(x) + x_img
        x2 = torch.sigmoid(self.conv3(img))
        x1 = x1*x2
        x1 = x1+x
        return x1, img

7.ORSNet(Original Resolution Subnetwork)

        为了保留输入图像的细节模型在最后一阶段引入了原始分辨率的子网(ORSNet:Original Resolution Subnetwork)。ORSNet不使用任何降采样操作并生成空间丰富的高分辨率特征。它由多个原始分辨率块BRB组成是模型的最后阶段结构如下

 图8

        可以看到输入为三个分别为上一层的输出和Stage2中的Decoder前后的特征。后两个输入每个输入又分为三个尺度三个尺度的通道数都先变成96然后在变成128小尺度的size都变成和大尺度一样最后做特征融合融合前会经过ORB(Original Resolution Block)模块。

        ORB由一连串的CAB组成还有一个大的残差边结构如下

 图9

        代码实现

# 位置MPRNet.py
## Original Resolution Block (ORB)
class ORB(nn.Module):
    def __init__(self, n_feat, kernel_size, reduction, act, bias, num_cab):
        super(ORB, self).__init__()
        modules_body = []
        modules_body = [CAB(n_feat, kernel_size, reduction, bias=bias, act=act) for _ in range(num_cab)]
        modules_body.append(conv(n_feat, n_feat, kernel_size))
        self.body = nn.Sequential(*modules_body)

    def forward(self, x):
        res = self.body(x)
        res += x
        return res

##########################################################################
class ORSNet(nn.Module):
    def __init__(self, n_feat, scale_orsnetfeats, kernel_size, reduction, act, bias, scale_unetfeats, num_cab):
        super(ORSNet, self).__init__()

        self.orb1 = ORB(n_feat+scale_orsnetfeats, kernel_size, reduction, act, bias, num_cab)
        self.orb2 = ORB(n_feat+scale_orsnetfeats, kernel_size, reduction, act, bias, num_cab)
        self.orb3 = ORB(n_feat+scale_orsnetfeats, kernel_size, reduction, act, bias, num_cab)

        self.up_enc1 = UpSample(n_feat, scale_unetfeats)
        self.up_dec1 = UpSample(n_feat, scale_unetfeats)

        self.up_enc2 = nn.Sequential(UpSample(n_feat+scale_unetfeats, scale_unetfeats), UpSample(n_feat, scale_unetfeats))
        self.up_dec2 = nn.Sequential(UpSample(n_feat+scale_unetfeats, scale_unetfeats), UpSample(n_feat, scale_unetfeats))

        self.conv_enc1 = nn.Conv2d(n_feat, n_feat+scale_orsnetfeats, kernel_size=1, bias=bias)
        self.conv_enc2 = nn.Conv2d(n_feat, n_feat+scale_orsnetfeats, kernel_size=1, bias=bias)
        self.conv_enc3 = nn.Conv2d(n_feat, n_feat+scale_orsnetfeats, kernel_size=1, bias=bias)

        self.conv_dec1 = nn.Conv2d(n_feat, n_feat+scale_orsnetfeats, kernel_size=1, bias=bias)
        self.conv_dec2 = nn.Conv2d(n_feat, n_feat+scale_orsnetfeats, kernel_size=1, bias=bias)
        self.conv_dec3 = nn.Conv2d(n_feat, n_feat+scale_orsnetfeats, kernel_size=1, bias=bias)

    def forward(self, x, encoder_outs, decoder_outs):
        x = self.orb1(x)
        x = x + self.conv_enc1(encoder_outs[0]) + self.conv_dec1(decoder_outs[0])

        x = self.orb2(x)
        x = x + self.conv_enc2(self.up_enc1(encoder_outs[1])) + self.conv_dec2(self.up_dec1(decoder_outs[1]))

        x = self.orb3(x)
        x = x + self.conv_enc3(self.up_enc2(encoder_outs[2])) + self.conv_dec3(self.up_dec2(decoder_outs[2]))

        return x

四、损失函数

        MPRNet主要使用了两个损失函数CharbonnierLoss和EdgeLoss公式如下

L=\sum_{S=1}^{3}[L_{char}(X_{S},Y)+\lambda L_{edge}(X_{S},Y)]

         其中累加是因为训练的时候三个阶段都有输出都需要个GT计算损失如图2的三个output该模型不是直接预测恢复的图像X_{S}而是预测残差图像R_{S}添加退化的输入图像I得到  X_{S}=I+R_{S}

        Deblurring和Deraining两个任务CharbonnierLoss和EdgeLoss做了加权求和比例1:0.05只使用了CharbonnierLoss我感觉是因为这里使用的噪声是某种分布入高斯分布、泊松分布的噪声不会引起剧烈的边缘差异所以Denoising没有使用EdgeLoss。

        下面简单介绍一下两种损失。

1.CharbonnierLoss

        公式如下

L_{char}=\sqrt{\left \| X_{s}-Y \right \|^{2}+\varepsilon^{2}}

        CharbonnierLoss在零点附近由于常数\varepsilon的存在梯度不会变成零避免梯度消失。函数曲线近似L1损失相比L2损失而言对异常值不敏感避免过分放大误差。

        代码实现

# 位置losses.py
class CharbonnierLoss(nn.Module):
    """Charbonnier Loss (L1)"""

    def __init__(self, eps=1e-3):
        super(CharbonnierLoss, self).__init__()
        self.eps = eps

    def forward(self, x, y):
        diff = x - y
        # loss = torch.sum(torch.sqrt(diff * diff + self.eps))
        loss = torch.mean(torch.sqrt((diff * diff) + (self.eps*self.eps)))
        return loss

2.EdgeLoss

        L1或者L2损失注重的是全局没有很好地考虑一些显著特征的影响 而显著的结构和纹理信息与人的主观感知效果高度相关是不能忽视的。

        边缘损失主要考虑纹理部分的差异可以很好地考虑高频的纹理结构信息 提高生成图像的细节表现公示如下

L_{edge}=\sqrt{\left \| \Delta (X_{S})-\Delta (Y) \right \|^{2}+\varepsilon^{2}}

        其中\Delta表示Laplacian边缘检测中的核函数\Delta (X_{S})表示对X_{S}做边缘检测公式中其他部分和CharbonnierLoss类似。

        代码实现

# 位置losses.py

class EdgeLoss(nn.Module):
    def __init__(self):
        super(EdgeLoss, self).__init__()
        k = torch.Tensor([[.05, .25, .4, .25, .05]])
        self.kernel = torch.matmul(k.t(),k).unsqueeze(0).repeat(3,1,1,1)
        if torch.cuda.is_available():
            self.kernel = self.kernel.cuda()
        self.loss = CharbonnierLoss()

    def conv_gauss(self, img):
        n_channels, _, kw, kh = self.kernel.shape
        img = F.pad(img, (kw//2, kh//2, kw//2, kh//2), mode='replicate')
        return F.conv2d(img, self.kernel, groups=n_channels)

    def laplacian_kernel(self, current):
        filtered    = self.conv_gauss(current)    # filter
        down        = filtered[:,:,::2,::2]               # downsample
        new_filter  = torch.zeros_like(filtered)
        new_filter[:,:,::2,::2] = down*4                  # upsample
        filtered    = self.conv_gauss(new_filter) # filter
        diff = current - filtered
        return diff

    def forward(self, x, y):
        loss = self.loss(self.laplacian_kernel(x), self.laplacian_kernel(y))
        return loss

        MPRNet的主要的内容就介绍到这主要是backbone的创新其他部分中规中矩关注不迷路。

        

阿里云国内75折 回扣 微信号：monov8
阿里云国际，腾讯云国际，低至75折。AWS 93折 免费开户实名账号 代冲值 优惠多多 微信号：monov8 飞机：@monov6
返回列表

上一篇：机器学习算法原理lightgbm、word2vec、cnn、lstm、textcnn、bert、transformer、随机森林、lr

下一篇：2373. 矩阵中的局部最大值

“【论文笔记】图像修复MPRNet:Multi-Stage Progressive Image Restoration 含代码解析” 的相关文章

公有云、私有云、混合云都是什么？有没有通俗一点的答案？企业选哪一种云比较好？1年前 (2023-02-02)
深入学习IO多路复用 select/poll/epoll 实现原理1年前 (2023-02-02)
使用动态输出打印内核的DEBUG信息1年前 (2023-02-02)
Typora软件与Markdown语法1年前 (2023-02-02)
【Linux】Linux进程的理解 --- 进程描述符、状态、优先级、切换…1年前 (2023-02-02)
洛谷 P1208混合牛奶 题解1年前 (2023-02-02)
区块链开发完整指南。如何开发一款区块链项目？1年前 (2023-02-02)
vue+vite+pinia+axios使用1年前 (2023-02-02)
ZooKeeper分布式协调服务(节点分类、事件监听、java集成)1年前 (2023-02-02)
Anaconda超详细安装教程（Windows环境下）1年前 (2023-02-02)

阿里云国际版