fcn8s.py

import torch
from torch import nn
from torchvision import models

from utils import get_upsampling_weight
#from .config import vgg16_path, vgg16_caffe_path


# This is implemented in full accordance with the original one (https://github.com/shelhamer/fcn.berkeleyvision.org)
class FCN8s(nn.Module):
    def __init__(self, num_classes, pretrained=True, caffe=False):
        super(FCN8s, self).__init__()
        vgg = models.vgg16(pretrained=True)
        if pretrained:
            vgg.load_state_dict(torch.load('/students/.torch/models/vgg16-397923af.pth'))
        features, classifier = list(vgg.features.children()), list(vgg.classifier.children())

        '''
        100 padding for 2 reasons:
            1) support very small input size
            2) allow cropping in order to match size of different layers' feature maps
        Note that the cropped part corresponds to a part of the 100 padding
        Spatial information of different layers' feature maps cannot be align exactly because of cropping, which is bad
        '''
        features[0].padding = (100, 100)

        for f in features:
            if 'MaxPool' in f.__class__.__name__:
                f.ceil_mode = True
            elif 'ReLU' in f.__class__.__name__:
                f.inplace = True

        self.features3 = nn.Sequential(*features[: 17])
        self.features4 = nn.Sequential(*features[17: 24])
        self.features5 = nn.Sequential(*features[24:])

        self.score_pool3 = nn.Conv2d(256, num_classes, kernel_size=1)
        self.score_pool4 = nn.Conv2d(512, num_classes, kernel_size=1)
        self.score_pool3.weight.data.zero_()
        self.score_pool3.bias.data.zero_()
        self.score_pool4.weight.data.zero_()
        self.score_pool4.bias.data.zero_()

        fc6 = nn.Conv2d(512, 4096, kernel_size=7)
        fc6.weight.data.copy_(classifier[0].weight.data.view(4096, 512, 7, 7))
        fc6.bias.data.copy_(classifier[0].bias.data)
        fc7 = nn.Conv2d(4096, 4096, kernel_size=1)
        fc7.weight.data.copy_(classifier[3].weight.data.view(4096, 4096, 1, 1))
        fc7.bias.data.copy_(classifier[3].bias.data)
        score_fr = nn.Conv2d(4096, num_classes, kernel_size=1)
        score_fr.weight.data.zero_()
        score_fr.bias.data.zero_()
        self.score_fr = nn.Sequential(
            fc6, nn.ReLU(inplace=True), nn.Dropout(), fc7, nn.ReLU(inplace=True), nn.Dropout(), score_fr
        )

        self.upscore2 = nn.ConvTranspose2d(num_classes, num_classes, kernel_size=4, stride=2, bias=False)
        self.upscore_pool4 = nn.ConvTranspose2d(num_classes, num_classes, kernel_size=4, stride=2, bias=False)
        self.upscore8 = nn.ConvTranspose2d(num_classes, num_classes, kernel_size=16, stride=8, bias=False)
        self.upscore2.weight.data.copy_(get_upsampling_weight(num_classes, num_classes, 4))
        self.upscore_pool4.weight.data.copy_(get_upsampling_weight(num_classes, num_classes, 4))
        self.upscore8.weight.data.copy_(get_upsampling_weight(num_classes, num_classes, 16))

    def forward(self, x):
        x_size = x.size()
        pool3 = self.features3(x)
        pool4 = self.features4(pool3)
        pool5 = self.features5(pool4)

        score_fr = self.score_fr(pool5)
        upscore2 = self.upscore2(score_fr)

        score_pool4 = self.score_pool4(0.01 * pool4)
        upscore_pool4 = self.upscore_pool4(score_pool4[:, :, 5: (5 + upscore2.size()[2]), 5: (5 + upscore2.size()[3])]
                                           + upscore2)

        score_pool3 = self.score_pool3(0.0001 * pool3)
        upscore8 = self.upscore8(score_pool3[:, :, 9: (9 + upscore_pool4.size()[2]), 9: (9 + upscore_pool4.size()[3])]
                                 + upscore_pool4)
        return upscore8[:, :, 31: (31 + x_size[2]), 31: (31 + x_size[3])].contiguous()