lib/pymafx/models/pose_resnet.py

# ------------------------------------------------------------------------------
# Copyright (c) Microsoft
# Licensed under the MIT License.
# Written by Bin Xiao ([email protected])
# ------------------------------------------------------------------------------

from __future__ import absolute_import, division, print_function

import logging
import os

import torch
import torch.nn as nn

BN_MOMENTUM = 0.1
logger = logging.getLogger(__name__)


def conv3x3(in_planes, out_planes, stride=1):
    """3x3 convolution with padding"""
    return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride, padding=1, bias=False)


class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, inplanes, planes, stride=1, downsample=None):
        super(BasicBlock, self).__init__()
        self.conv1 = conv3x3(inplanes, planes, stride)
        self.bn1 = nn.BatchNorm2d(planes, momentum=BN_MOMENTUM)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3(planes, planes)
        self.bn2 = nn.BatchNorm2d(planes, momentum=BN_MOMENTUM)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):
        residual = x

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

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

        if self.downsample is not None:
            residual = self.downsample(x)

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

        return out


class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, inplanes, planes, stride=1, downsample=None):
        super(Bottleneck, self).__init__()
        self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
        self.bn1 = nn.BatchNorm2d(planes, momentum=BN_MOMENTUM)
        self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride, padding=1, bias=False)
        self.bn2 = nn.BatchNorm2d(planes, momentum=BN_MOMENTUM)
        self.conv3 = nn.Conv2d(planes, planes * self.expansion, kernel_size=1, bias=False)
        self.bn3 = nn.BatchNorm2d(planes * self.expansion, momentum=BN_MOMENTUM)
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):
        residual = x

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

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

        out = self.conv3(out)
        out = self.bn3(out)

        if self.downsample is not None:
            residual = self.downsample(x)

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

        return out


class PoseResNet(nn.Module):
    def __init__(self, block, layers, cfg, global_mode, **kwargs):
        self.inplanes = 64
        extra = cfg.POSE_RES_MODEL.EXTRA
        self.extra = extra
        self.deconv_with_bias = extra.DECONV_WITH_BIAS

        super(PoseResNet, self).__init__()
        self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, bias=False)
        self.bn1 = nn.BatchNorm2d(64, momentum=BN_MOMENTUM)
        self.relu = nn.ReLU(inplace=True)
        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        self.layer1 = self._make_layer(block, 64, layers[0])
        self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
        self.layer3 = self._make_layer(block, 256, layers[2], stride=2)
        self.layer4 = self._make_layer(block, 512, layers[3], stride=2)

        self.global_mode = global_mode
        if self.global_mode:
            self.avgpool = nn.AvgPool2d(7, stride=1)
            self.deconv_layers = None
        else:
            # used for deconv layers
            self.deconv_layers = self._make_deconv_layer(
                extra.NUM_DECONV_LAYERS,
                extra.NUM_DECONV_FILTERS,
                extra.NUM_DECONV_KERNELS,
            )

        # self.final_layer = nn.Conv2d(
        #     in_channels=extra.NUM_DECONV_FILTERS[-1],
        #     out_channels=17,
        #     kernel_size=extra.FINAL_CONV_KERNEL,
        #     stride=1,
        #     padding=1 if extra.FINAL_CONV_KERNEL == 3 else 0
        # )
        self.final_layer = None

    def _make_layer(self, block, planes, blocks, stride=1):
        downsample = None
        if stride != 1 or self.inplanes != planes * block.expansion:
            downsample = nn.Sequential(
                nn.Conv2d(
                    self.inplanes,
                    planes * block.expansion,
                    kernel_size=1,
                    stride=stride,
                    bias=False
                ),
                nn.BatchNorm2d(planes * block.expansion, momentum=BN_MOMENTUM),
            )

        layers = []
        layers.append(block(self.inplanes, planes, stride, downsample))
        self.inplanes = planes * block.expansion
        for i in range(1, blocks):
            layers.append(block(self.inplanes, planes))

        return nn.Sequential(*layers)

    def _get_deconv_cfg(self, deconv_kernel, index):
        if deconv_kernel == 4:
            padding = 1
            output_padding = 0
        elif deconv_kernel == 3:
            padding = 1
            output_padding = 1
        elif deconv_kernel == 2:
            padding = 0
            output_padding = 0

        return deconv_kernel, padding, output_padding

    def _make_deconv_layer(self, num_layers, num_filters, num_kernels):
        assert num_layers == len(num_filters), \
            'ERROR: num_deconv_layers is different len(num_deconv_filters)'
        assert num_layers == len(num_kernels), \
            'ERROR: num_deconv_layers is different len(num_deconv_filters)'

        layers = []
        for i in range(num_layers):
            kernel, padding, output_padding = \
                self._get_deconv_cfg(num_kernels[i], i)

            planes = num_filters[i]
            layers.append(
                nn.ConvTranspose2d(
                    in_channels=self.inplanes,
                    out_channels=planes,
                    kernel_size=kernel,
                    stride=2,
                    padding=padding,
                    output_padding=output_padding,
                    bias=self.deconv_with_bias
                )
            )
            layers.append(nn.BatchNorm2d(planes, momentum=BN_MOMENTUM))
            layers.append(nn.ReLU(inplace=True))
            self.inplanes = planes

        return nn.Sequential(*layers)

    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)
        x = self.maxpool(x)

        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)

        # x = self.deconv_layers(x)
        # x = self.final_layer(x)

        if self.global_mode:
            g_feat = self.avgpool(x)
            g_feat = g_feat.view(g_feat.size(0), -1)
            s_feat_list = [g_feat]
        else:
            g_feat = None
            if self.extra.NUM_DECONV_LAYERS == 3:
                deconv_blocks = [
                    self.deconv_layers[0:3], self.deconv_layers[3:6], self.deconv_layers[6:9]
                ]

            s_feat_list = []
            s_feat = x
            for i in range(self.extra.NUM_DECONV_LAYERS):
                s_feat = deconv_blocks[i](s_feat)
                s_feat_list.append(s_feat)

        return s_feat_list, g_feat

    def init_weights(self, pretrained=''):
        if os.path.isfile(pretrained):
            # logger.info('=> init deconv weights from normal distribution')
            if self.deconv_layers is not None:
                for name, m in self.deconv_layers.named_modules():
                    if isinstance(m, nn.ConvTranspose2d):
                        # logger.info('=> init {}.weight as normal(0, 0.001)'.format(name))
                        # logger.info('=> init {}.bias as 0'.format(name))
                        nn.init.normal_(m.weight, std=0.001)
                        if self.deconv_with_bias:
                            nn.init.constant_(m.bias, 0)
                    elif isinstance(m, nn.BatchNorm2d):
                        # logger.info('=> init {}.weight as 1'.format(name))
                        # logger.info('=> init {}.bias as 0'.format(name))
                        nn.init.constant_(m.weight, 1)
                        nn.init.constant_(m.bias, 0)
            if self.final_layer is not None:
                logger.info('=> init final conv weights from normal distribution')
                for m in self.final_layer.modules():
                    if isinstance(m, nn.Conv2d):
                        # nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
                        logger.info('=> init {}.weight as normal(0, 0.001)'.format(name))
                        logger.info('=> init {}.bias as 0'.format(name))
                        nn.init.normal_(m.weight, std=0.001)
                        nn.init.constant_(m.bias, 0)

            pretrained_state_dict = torch.load(pretrained)
            logger.info('=> loading pretrained model {}'.format(pretrained))
            self.load_state_dict(pretrained_state_dict, strict=False)
        elif pretrained:
            logger.error('=> please download pre-trained models first!')
            raise ValueError('{} is not exist!'.format(pretrained))
        else:
            logger.info('=> init weights from normal distribution')
            for m in self.modules():
                if isinstance(m, nn.Conv2d):
                    # nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
                    nn.init.normal_(m.weight, std=0.001)
                    # nn.init.constant_(m.bias, 0)
                elif isinstance(m, nn.BatchNorm2d):
                    nn.init.constant_(m.weight, 1)
                    nn.init.constant_(m.bias, 0)
                elif isinstance(m, nn.ConvTranspose2d):
                    nn.init.normal_(m.weight, std=0.001)
                    if self.deconv_with_bias:
                        nn.init.constant_(m.bias, 0)


resnet_spec = {
    18: (BasicBlock, [2, 2, 2, 2]), 34: (BasicBlock, [3, 4, 6, 3]), 50: (Bottleneck, [3, 4, 6, 3]),
    101: (Bottleneck, [3, 4, 23, 3]), 152: (Bottleneck, [3, 8, 36, 3])
}


def get_resnet_encoder(cfg, init_weight=True, global_mode=False, **kwargs):
    num_layers = cfg.POSE_RES_MODEL.EXTRA.NUM_LAYERS

    block_class, layers = resnet_spec[num_layers]

    model = PoseResNet(block_class, layers, cfg, global_mode, **kwargs)

    if init_weight:
        if num_layers == 50:
            if cfg.POSE_RES_MODEL.PRETR_SET in ['imagenet']:
                model.init_weights(cfg.POSE_RES_MODEL.PRETRAINED_IM)
                logger.info('loaded ResNet imagenet pretrained model')
            elif cfg.POSE_RES_MODEL.PRETR_SET in ['coco']:
                model.init_weights(cfg.POSE_RES_MODEL.PRETRAINED_COCO)
                logger.info('loaded ResNet coco pretrained model')
        else:
            raise NotImplementedError

    return model