fpn网络结构及代码

---

一、fpn网络结构

resnet结构组成。

fpn（Feature Pyramid Networks）架构如上图所示，以resnet50作为backbone。通过四部分的卷积，得到C2-C5特征图。然后统一使用1x1卷积层来保持特征图的维度一致。从最上层开始，每层进行上采用过程，采用的方法是‘nearnest’，把上采样的结果与当前层的特征图进行拼接得到当前层的特征图。

fpn示意图，要在不同特征层上进行预测，论文中采用的是上述公式。k0初始值设为4，比如当前图像的高宽为112，则计算得到的k=3，在C3特征层上进行预测。

二、代码

class Decoder(nn.Module):
    def __init__(self, C3_size, C4_size, C5_size, feature_size=256):
        super(Decoder, self).__init__()

        # upsample C5 to get P5 from the FPN paper
        self.P5_1 = nn.Conv2d(C5_size, feature_size, kernel_size=1, stride=1, padding=0)
        self.P5_upsampled = nn.Upsample(scale_factor=2, mode='nearest')
        self.P5_2 = nn.Conv2d(feature_size, feature_size, kernel_size=3, stride=1, padding=1)

        # add P5 elementwise to C4
        self.P4_1 = nn.Conv2d(C4_size, feature_size, kernel_size=1, stride=1, padding=0)
        self.P4_upsampled = nn.Upsample(scale_factor=2, mode='nearest')
        self.P4_2 = nn.Conv2d(feature_size, feature_size, kernel_size=3, stride=1, padding=1)

        # add P4 elementwise to C3
        self.P3_1 = nn.Conv2d(C3_size, feature_size, kernel_size=1, stride=1, padding=0)
        self.P3_upsampled = nn.Upsample(scale_factor=2, mode='nearest')
        self.P3_2 = nn.Conv2d(feature_size, feature_size, kernel_size=3, stride=1, padding=1)


    def forward(self, inputs):
        C3, C4, C5 = inputs

        P5_x = self.P5_1(C5)
        P5_upsampled_x = self.P5_upsampled(P5_x)
        P5_x = self.P5_2(P5_x)

        P4_x = self.P4_1(C4)
        P4_x = P5_upsampled_x + P4_x
        P4_upsampled_x = self.P4_upsampled(P4_x)
        P4_x = self.P4_2(P4_x)

        P3_x = self.P3_1(C3)
        P3_x = P3_x + P4_upsampled_x
        P3_x = self.P3_2(P3_x)

        return [P3_x, P4_x, P5_x]