目錄

Mobilenetv2的改進(jìn)

淺層特征和深層特征的融合

完整代碼

參考資料

Mobilenetv2的改進(jìn)

在DeeplabV3當(dāng)中，一般不會5次下采樣，可選的有3次下采樣和4次下采樣。因?yàn)橐M(jìn)行五次下采樣的話會損失較多的信息。

在這里mobilenetv2會從之前寫好的模塊中得到，但注意的是，我們在這里獲得的特征是[-1]，也就是最后的1x1卷積不取，只取循環(huán)完后的模型。

down_idx是InvertedResidual進(jìn)行的次數(shù)。

# t, c, n, s
[1, 16, 1, 1],?
[6, 24, 2, 2], ?? 2
[6, 32, 3, 2], ? ?4
[6, 64, 4, 2], ? ?7 ?
[6, 96, 3, 1],
[6, 160, 3, 2], ??14
[6, 320, 1, 1],?

根據(jù)下采樣的不同，當(dāng)downsample_factor=8時(shí)，進(jìn)行3次下采樣，對倒數(shù)兩次，步長為2的InvertedResidual進(jìn)行參數(shù)的修改，讓步長變?yōu)?，膨脹系數(shù)為2。

當(dāng)downsample_factor=16時(shí)，進(jìn)行4次下采樣，只需對最后一次進(jìn)行參數(shù)的修改。

import torch
import torch.nn as nn
import torch.nn.functional as F
from functools import partial

from net.mobilenetv2 import mobilenetv2
from net.ASPP import ASPP

class MobileNetV2(nn.Module):
    def __init__(self, downsample_factor=8, pretrained=True):
        super(MobileNetV2, self).__init__()
        
        model           = mobilenetv2(pretrained)
        self.features   = model.features[:-1]

        self.total_idx  = len(self.features)
        self.down_idx   = [2, 4, 7, 14]

        if downsample_factor == 8:
            for i in range(self.down_idx[-2], self.down_idx[-1]):
                self.features[i].apply(
                    partial(self._nostride_dilate, dilate=2)
                )
            for i in range(self.down_idx[-1], self.total_idx):
                self.features[i].apply(
                    partial(self._nostride_dilate, dilate=4)
                )
        elif downsample_factor == 16:
            for i in range(self.down_idx[-1], self.total_idx):
                self.features[i].apply(
                    partial(self._nostride_dilate, dilate=2)
                )
        
    def _nostride_dilate(self, m, dilate):
        classname = m.__class__.__name__
        if classname.find('Conv') != -1:
            if m.stride == (2, 2):
                m.stride = (1, 1)
                if m.kernel_size == (3, 3):
                    m.dilation = (dilate//2, dilate//2)
                    m.padding = (dilate//2, dilate//2)
            else:
                if m.kernel_size == (3, 3):
                    m.dilation = (dilate, dilate)
                    m.padding = (dilate, dilate)

    def forward(self, x):
        low_level_features = self.features[:4](x)
        x = self.features[4:](low_level_features)
        return low_level_features, x

forward當(dāng)中，會輸出兩個(gè)特征層，一個(gè)是淺層特征層，具有淺層的語義信息；另一個(gè)是深層特征層，具有深層的語義信息。

淺層特征和深層特征的融合

?具有高語義信息的部分先進(jìn)行上采樣，低語義信息的特征層進(jìn)行1x1卷積，二者進(jìn)行特征融合，再進(jìn)行3x3卷積進(jìn)行特征提取

self.aspp = ASPP(dim_in=in_channels, dim_out=256, rate=16//downsample_factor)

這一步就是獲得那個(gè)綠色的特征層；

low_level_features = self.shortcut_conv(low_level_features)

從這里將是對淺層特征的初步處理（1x1卷積）；

x = F.interpolate(x, size=(low_level_features.size(2), low_level_features.size(3)), mode='bilinear', align_corners=True)
x = self.cat_conv(torch.cat((x, low_level_features), dim=1))

上采樣后進(jìn)行特征融合，這樣我們輸入和輸出的大小才相同，每一個(gè)像素點(diǎn)才能進(jìn)行預(yù)測；

完整代碼

# deeplabv3plus.py

import torch
import torch.nn as nn
import torch.nn.functional as F
from functools import partial

from net.xception import xception
from net.mobilenetv2 import mobilenetv2
from net.ASPP import ASPP

class MobileNetV2(nn.Module):
    def __init__(self, downsample_factor=8, pretrained=True):
        super(MobileNetV2, self).__init__()
        
        model           = mobilenetv2(pretrained)
        self.features   = model.features[:-1]

        self.total_idx  = len(self.features)
        self.down_idx   = [2, 4, 7, 14]

        if downsample_factor == 8:
            for i in range(self.down_idx[-2], self.down_idx[-1]):
                self.features[i].apply(
                    partial(self._nostride_dilate, dilate=2)
                )
            for i in range(self.down_idx[-1], self.total_idx):
                self.features[i].apply(
                    partial(self._nostride_dilate, dilate=4)
                )
        elif downsample_factor == 16:
            for i in range(self.down_idx[-1], self.total_idx):
                self.features[i].apply(
                    partial(self._nostride_dilate, dilate=2)
                )
        
    def _nostride_dilate(self, m, dilate):
        classname = m.__class__.__name__
        if classname.find('Conv') != -1:
            if m.stride == (2, 2):
                m.stride = (1, 1)
                if m.kernel_size == (3, 3):
                    m.dilation = (dilate//2, dilate//2)
                    m.padding = (dilate//2, dilate//2)
            else:
                if m.kernel_size == (3, 3):
                    m.dilation = (dilate, dilate)
                    m.padding = (dilate, dilate)

    def forward(self, x):
        low_level_features = self.features[:4](x)
        x = self.features[4:](low_level_features)
        return low_level_features, x

class DeepLab(nn.Module):
    def __init__(self, num_classes, backbone="mobilenet", pretrained=True, downsample_factor=16):
        super(DeepLab, self).__init__()
        if backbone=="xception":
         
            #   獲得兩個(gè)特征層：淺層特征 主干部分    
            self.backbone = xception(downsample_factor=downsample_factor, pretrained=pretrained)
            in_channels = 2048
            low_level_channels = 256
        elif backbone=="mobilenet":

            #   獲得兩個(gè)特征層：淺層特征 主干部分
            self.backbone = MobileNetV2(downsample_factor=downsample_factor, pretrained=pretrained)
            in_channels = 320
            low_level_channels = 24
        else:
            raise ValueError('Unsupported backbone - `{}`, Use mobilenet, xception.'.format(backbone))

        #   ASPP特征提取模塊
        #   利用不同膨脹率的膨脹卷積進(jìn)行特征提取
        self.aspp = ASPP(dim_in=in_channels, dim_out=256, rate=16//downsample_factor)
       
        # 淺層特征邊
        self.shortcut_conv = nn.Sequential(
            nn.Conv2d(low_level_channels, 48, 1),
            nn.BatchNorm2d(48),
            nn.ReLU(inplace=True)
        )		

        self.cat_conv = nn.Sequential(
            nn.Conv2d(48+256, 256, kernel_size=(3,3), stride=(1,1), padding=1),
            nn.BatchNorm2d(256),
            nn.ReLU(inplace=True),
            nn.Dropout(0.5),

            nn.Conv2d(256, 256, kernel_size=(3,3), stride=(1,1), padding=1),
            nn.BatchNorm2d(256),
            nn.ReLU(inplace=True),

            nn.Dropout(0.1),
        )
        self.cls_conv = nn.Conv2d(256, num_classes, kernel_size=(1,1), stride=(1,1))

    def forward(self, x):
        H, W = x.size(2), x.size(3)

        # 獲得兩個(gè)特征層，low_level_features: 淺層特征-進(jìn)行卷積處理
        #                x : 主干部分-利用ASPP結(jié)構(gòu)進(jìn)行加強(qiáng)特征提取
  
        low_level_features, x = self.backbone(x)
        x = self.aspp(x)
        low_level_features = self.shortcut_conv(low_level_features)

        #   將加強(qiáng)特征邊上采樣，與淺層特征堆疊后利用卷積進(jìn)行特征提取
        x = F.interpolate(x, size=(low_level_features.size(2), low_level_features.size(3)), mode='bilinear', align_corners=True)
        x = self.cat_conv(torch.cat((x, low_level_features), dim=1))
        x = self.cls_conv(x)
        x = F.interpolate(x, size=(H, W), mode='bilinear', align_corners=True)
        return x

參考資料

DeepLabV3-/論文精選 at main · Auorui/DeepLabV3- (github.com)

(6條消息) 憨批的語義分割重制版9——Pytorch 搭建自己的DeeplabV3+語義分割平臺_Bubbliiiing的博客-CSDN博客文章來源地址http://www.zghlxwxcb.cn/news/detail-805206.html

到了這里，關(guān)于DeepLabV3+：Mobilenetv2的改進(jìn)以及淺層特征和深層特征的融合的文章就介紹完了。如果您還想了解更多內(nèi)容，請?jiān)谟疑辖撬阉鱐OY模板網(wǎng)以前的文章或繼續(xù)瀏覽下面的相關(guān)文章，希望大家以后多多支持TOY模板網(wǎng)！