48 全连接卷积神经网络 FCN [动手学深度学习v2]

1. FCN是用深度神经网络来做语义分割的奠基性工作；它用转置卷积层来替换CNN最后的全连接层，从而可以实现每个像素的预测。

2. 初始化转置卷积层

   # 利用双线性插值来初始化卷积核参数
   def bilinear_kernel(in_channels, out_channels, kernel_size):
       factor = (kernel_size + 1) // 2
       if kernel_size % 2 == 1:
           center = factor - 1
       else:
           center = factor - 0.5
       og = (torch.arange(kernel_size).reshape(-1, 1), torch.arange(kernel_size).reshape(1, -1))
       filt = (1 - torch.abs(og[0] - center) / factor) * (1 - torch.abs(og[1] - center) / factor)
       weight = torch.zeros((in_channels, out_channels, kernel_size, kernel_size))
       weight[range(in_channels), range(out_channels), :, :] = filt
       return weight