基于pytorch和torchvision实现DeformableConv

相关代码如下：

import torch
import torchvision.ops
from torch import nn


class DeformableConv2d(nn.Module):
    def __init__(self,
                 in_channels,
                 out_channels,
                 kernel_size=3,
                 stride=1,
                 padding=1,
                 bias=False):
        super(DeformableConv2d, self).__init__()

        self.stride = stride if type(stride) == tuple else (stride, stride)
        self.padding = padding

        self.offset_conv = nn.Conv2d(in_channels,
                                     2 * kernel_size * kernel_size,
                                     kernel_size=kernel_size,
                                     stride=stride,
                                     padding=self.padding,
                                     bias=True)

        nn.init.constant_(self.offset_conv.weight, 0.)
        nn.init.constant_(self.offset_conv.bias, 0.)

        self.modulator_conv = nn.Conv2d(in_channels,
                                        1 * kernel_size * kernel_size,
                                        kernel_size=kernel_size,
                                        stride=stride,
                                        padding=self.padding,
                                        bias=True)

        nn.init.constant_(self.modulator_conv.weight, 0.)
        nn.init.constant_(self.modulator_conv.bias, 0.)

        self.regular_conv = nn.Conv2d(in_channels=in_channels,
                                      out_channels=out_channels,
                                      kernel_size=kernel_size,
                                      stride=stride,
                                      padding=self.padding,
                                      bias=bias)

    def forward(self, x):
        # h, w = x.shape[2:]
        # max_offset = max(h, w)/4.

        offset = self.offset_conv(x)  # .clamp(-max_offset, max_offset)
        # modulator = 2. * torch.sigmoid(self.modulator_conv(x))

        # self.regular_conv.weight = self.regular_conv.weight.half() if x.dtype == torch.float16 else \
        #     self.regular_conv.weight
        x = torchvision.ops.deform_conv2d(input=x.float(),
                                          offset=offset.float(),
                                          weight=self.regular_conv.weight,
                                          bias=self.regular_conv.bias,
                                          padding=(self.padding, self.padding),
                                          # mask=modulator,
                                          stride=self.stride,
                                          )
        return x