【技术文档】centernet(姿态估计)

模型结构

backbone dla34

dla（Deep Layer Aggregation）
We introduce two structures for deep layer aggregation (DLA): iterative deep aggrega-
tion (IDA) and hierarchical deep aggregation (HDA).
Hierarchical deep aggregation merges blocks and stages in a tree to preserve and combine feature channels.
我们介绍两种结构深层聚合(DLA):迭代深层聚合 (IDA)和层次深度聚合(HDA)。
IDA focuses on fusing resolutions and scales while HDA focuses on merging features from all modules and channels.
IDA主要负责不同空间尺度信息的融合，HDA侧重于合并来自所有模块和通道的特性。
HDA

基本卷积结构block

class BasicBlock(nn.Module):
    def __init__(self, inplanes, planes, stride=1, dilation=1):
        super(BasicBlock, self).__init__()
        self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3,
                               stride=stride, padding=dilation,
                               bias=False, dilation=dilation)
        self.bn1 = nn.BatchNorm2d(planes, momentum=BN_MOMENTUM)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = nn.Conv2d(planes, planes, kernel_size=3,
                               stride=1, padding=dilation,
                               bias=False, dilation=dilation)
        self.bn2 = nn.BatchNorm2d(planes, momentum=BN_MOMENTUM)
        self.stride = stride

    def forward(self, x, residual=None):
        if residual is None:
            residual = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)

        out += residual
        out = self.relu(out)

        return out

构成具备树形结构的模块

class Root(nn.Module):
    def __init__(self, in_channels, out_channels, kernel_size, residual):
        super(Root, self).__init__()
        self.conv = nn.Conv2d(
            in_channels, out_channels, 1,
            stride=1, bias=False, padding=(kernel_size - 1) // 2)
        self.bn = nn.BatchNorm2d(out_channels, momentum=BN_MOMENTUM)
        self.relu = nn.ReLU(inplace=True)
        self.residual = residual

    def forward(self, *x):
        children = x
        x = self.conv(torch.cat(x, 1))
        x = self.bn(x)
        if self.residual:
            x += children[0]
        x = self.relu(x)

        return x


class Tree(nn.Module):
    def __init__(self, levels, block, in_channels, out_channels, stride=1,
                 level_root=False, root_dim=0, root_kernel_size=1,
                 dilation=1, root_residual=False):
        super(Tree, self).__init__()
        if root_dim == 0:
            root_dim = 2 * out_channels
        if level_root:
            root_dim += in_channels
        if levels == 1:
            self.tree1 = block(in_channels, out_channels, stride,
                               dilation=dilation)
            self.tree2 = block(out_channels, out_channels, 1,
                               dilation=dilation)
        else:
            self.tree1 = Tree(levels - 1, block, in_channels, out_channels,
                              stride, root_dim=0,
                              root_kernel_size=root_kernel_size,
                              dilation=dilation, root_residual=root_residual)
            self.tree2 = Tree(levels - 1, block, out_channels, out_channels,
                              root_dim=root_dim + out_channels,
                              root_kernel_size=root_kernel_size,
                              dilation=dilation, root_residual=root_residual)
        if levels == 1:
            self.root = Root(root_dim, out_channels, root_kernel_size,
                             root_residual)
        self.level_root = level_root
        self.root_dim = root_dim
        self.downsample = None
        self.project = None
        self.levels = levels
        if stride > 1:
            self.downsample = nn.MaxPool2d(stride, stride=stride)
        if in_channels != out_channels:
            self.project = nn.Sequential(
                nn.Conv2d(in_channels, out_channels,
                          kernel_size=1, stride=1, bias=False),
                nn.BatchNorm2d(out_channels, momentum=BN_MOMENTUM)
            )

    def forward(self, x, residual=None, children=None):
        children = [] if children is None else children
        bottom = self.downsample(x) if self.downsample else x
        residual = self.project(bottom) if self.project else bottom
        if self.level_root:
            children.append(bottom)
        x1 = self.tree1(x, residual)
        if self.levels == 1:
            x2 = self.tree2(x1)
            x = self.root(x2, x1, *children)
        else:
            children.append(x1)
            x = self.tree2(x1, children=children)
        return x

HDA
IDA

IDAUp
IDAUp是IDA模块需要不断重复使用的操作，每一次操作都需要在上此次操作结果的基础上。如上图所示，每一层的结果都是由IDAUp生成，连续使用就形成了现在的金字塔结构

class IDAUp(nn.Module):

    def __init__(self, o, channels, up_f):
        super(IDAUp, self).__init__()
        for i in range(1, len(channels)):
            c = channels[i]
            f = int(up_f[i])  
            proj = DeformConv(c, o)
            node = DeformConv(o, o)
     
            up = nn.ConvTranspose2d(o, o, f * 2, stride=f, 
                                    padding=f // 2, output_padding=0,
                                    groups=o, bias=False)
            fill_up_weights(up)

            setattr(self, 'proj_' + str(i), proj)
            setattr(self, 'up_' + str(i), up)
            setattr(self, 'node_' + str(i), node)
                 
        
    def forward(self, layers, startp, endp):
        for i in range(startp + 1, endp):
            upsample = getattr(self, 'up_' + str(i - startp))
            project = getattr(self, 'proj_' + str(i - startp))
            layers[i] = upsample(project(layers[i]))
            node = getattr(self, 'node_' + str(i - startp))
            layers[i] = node(layers[i] + layers[i - 1])

class DLAUp(nn.Module):
    def __init__(self, startp, channels, scales, in_channels=None):
        super(DLAUp, self).__init__()
        self.startp = startp
        if in_channels is None:
            in_channels = channels
        self.channels = channels
        channels = list(channels)
        scales = np.array(scales, dtype=int)
        for i in range(len(channels) - 1):
            j = -i - 2
            setattr(self, 'ida_{}'.format(i),
                    IDAUp(channels[j], in_channels[j:],
                          scales[j:] // scales[j]))
            scales[j + 1:] = scales[j]
            in_channels[j + 1:] = [channels[j] for _ in channels[j + 1:]]

    def forward(self, layers):
        out = [layers[-1]] # start with 32
        for i in range(len(layers) - self.startp - 1):
            ida = getattr(self, 'ida_{}'.format(i