def __init__(self, stride=16):
super(Net, self).__init__()
if stride == 16:
self.resnet50 = resnet50.resnet50(pretrained=True, strides=(2, 2, 2, 1))
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1, self.resnet50.relu, self.resnet50.maxpool,
self.resnet50.layer1)
else:
self.resnet50 = resnet50.resnet50(pretrained=True, strides=(2, 2, 1, 1), dilations=(1, 1, 2, 2))
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1, self.resnet50.relu, self.resnet50.maxpool,
self.resnet50.layer1)
self.stage2 = nn.Sequential(self.resnet50.layer2)
self.stage3 = nn.Sequential(self.resnet50.layer3)
self.stage4 = nn.Sequential(self.resnet50.layer4)
self.classifier = nn.Conv2d(2048, 20, 1, bias=False)
self.backbone = nn.ModuleList([self.stage1, self.stage2, self.stage3, self.stage4])
self.newly_added = nn.ModuleList([self.classifier])
def __init__(self):
super(Net, self).__init__()
self.resnet50 = resnet50.resnet50(pretrained=True,
strides=[2, 2, 2, 1])
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1,
self.resnet50.relu, self.resnet50.maxpool)
self.stage2 = nn.Sequential(self.resnet50.layer1)
self.stage3 = nn.Sequential(self.resnet50.layer2)
self.stage4 = nn.Sequential(self.resnet50.layer3)
self.stage5 = nn.Sequential(self.resnet50.layer4)
# branch: class boundary detection
self.fc_edge1 = nn.Sequential(
nn.Conv2d(64, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.ReLU(inplace=True),
)
self.fc_edge2 = nn.Sequential(
nn.Conv2d(256, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.ReLU(inplace=True),
)
self.fc_edge3 = nn.Sequential(
nn.Conv2d(512, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.Upsample(scale_factor=2, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_edge4 = nn.Sequential(
nn.Conv2d(1024, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.Upsample(scale_factor=4, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_edge5 = nn.Sequential(
nn.Conv2d(2048, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.Upsample(scale_factor=4, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_edge6 = nn.Conv2d(160, 1, 1, bias=True)
self.backbone = nn.ModuleList(
[self.stage1, self.stage2, self.stage3, self.stage4, self.stage5])
self.edge_layers = nn.ModuleList([
self.fc_edge1, self.fc_edge2, self.fc_edge3, self.fc_edge4,
self.fc_edge5, self.fc_edge6
])
self.classifier = nn.Conv2d(in_channels=2048,
out_channels=20,
kernel_size=1,
bias=False)
def __init__(self, model_dir, num_classes):
super(Net, self).__init__()
self.num_classes = num_classes
self.resnet50 = resnet50.resnet50(pretrained=True,
strides=(2, 2, 2, 1))
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1,
self.resnet50.relu, self.resnet50.maxpool,
self.resnet50.layer1)
self.stage2 = nn.Sequential(self.resnet50.layer2)
self.stage3 = nn.Sequential(self.resnet50.layer3)
self.stage4 = nn.Sequential(self.resnet50.layer4)
self.classifier = nn.Conv2d(2048, self.num_classes, 1, bias=False)
self.backbone = nn.ModuleList(
[self.stage1, self.stage2, self.stage3, self.stage4])
self.newly_added = nn.ModuleList([self.classifier])
def __init__(self):
super(Net, self).__init__()
self.resnet50 = resnet50.resnet50(pretrained=False,
strides=[2, 2, 2, 1])
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1,
self.resnet50.relu, self.resnet50.maxpool,
self.resnet50.layer1)
self.stage2 = nn.Sequential(self.resnet50.layer2)
self.stage3 = nn.Sequential(self.resnet50.layer3)
self.stage4 = nn.Sequential(self.resnet50.layer4)
self.classifier = nn.Conv2d(in_channels=2048,
out_channels=20,
kernel_size=1,
bias=False)
self.backbone = nn.ModuleList(
[self.stage1, self.stage2, self.stage3, self.stage4])
self.newly_added = nn.ModuleList([self.classifier])
def __init__(self,class_number=5, fine_tune=True):
super().__init__()
encoder = resnet50(pretrained=fine_tune)
self.start = nn.Sequential(encoder.conv1,encoder.bn1,
encoder.relu)
self.maxpool = encoder.maxpool
self.low_feature = nn.Sequential(nn.Conv2d(64,48,1,1),nn.ReLU(inplace=True)) # no bn, has bias and relu
self.layer1 = encoder.layer1#256
self.layer2 = encoder.layer2#512
self.layer3 = encoder.layer3#1024
self.layer4 = encoder.layer4#2048
self.aspp = ASPP(in_channel=2048, depth=256)
self.conv_cat = nn.Sequential(nn.Conv2d(256+48,256,3,1,padding=1),nn.ReLU(inplace=True))
self.conv_cat1 = nn.Sequential(nn.Conv2d(256,256,3,1,padding=1),nn.ReLU(inplace=True))
self.conv_cat2 = nn.Sequential(nn.Conv2d(256,256,3,1,padding=1),nn.ReLU(inplace=True))
self.score = nn.Conv2d(256,class_number,1,1)# no relu and first conv then upsample, reduce memory
def __init__(self):
super(Net, self).__init__()
self.resnet50 = resnet50.resnet50(pretrained=True,
strides=(2, 2, 2, 1))
state_dict = torch.load("../temp_data/resnet50-19c8e357.pth")
state_dict.pop('fc.weight')
state_dict.pop('fc.bias')
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1,
self.resnet50.relu, self.resnet50.maxpool,
self.resnet50.layer1)
self.stage2 = nn.Sequential(self.resnet50.layer2)
self.stage3 = nn.Sequential(self.resnet50.layer3)
self.stage4 = nn.Sequential(self.resnet50.layer4)
self.classifier = nn.Conv2d(2048, 2, 1, bias=False)
self.backbone = nn.ModuleList(
[self.stage1, self.stage2, self.stage3, self.stage4])
self.newly_added = nn.ModuleList([self.classifier])
def __init__(self, n_classes=20, in_channels=3, pretrained=True):
super(Net, self).__init__()
self.n_classes = n_classes
self.in_channels = in_channels
self.pretrained = pretrained
self.resnet50 = resnet50.resnet50(pretrained=pretrained,
in_channels=in_channels,
strides=(2, 2, 2, 1))
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1,
self.resnet50.relu, self.resnet50.maxpool,
self.resnet50.layer1)
self.stage2 = nn.Sequential(self.resnet50.layer2)
self.stage3 = nn.Sequential(self.resnet50.layer3)
self.stage4 = nn.Sequential(self.resnet50.layer4)
self.classifier = nn.Conv2d(2048, self.n_classes, 1, bias=False)
self.backbone = nn.ModuleList(
[self.stage1, self.stage2, self.stage3, self.stage4])
self.newly_added = nn.ModuleList([self.classifier])
def __init__(self):
super(CamNet, self).__init__()
self.resnet50 = resnet50.resnet50(pretrained=True,
strides=(2, 2, 2, 1))
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1,
self.resnet50.relu, self.resnet50.maxpool,
self.resnet50.layer1)
self.stage2 = nn.Sequential(self.resnet50.layer2)
self.stage3 = nn.Sequential(self.resnet50.layer3)
self.stage4 = nn.Sequential(self.resnet50.layer4)
self.classifier = nn.Conv2d(2048, n_classes, 1, bias=False)
self.backbone = nn.ModuleList(
[self.stage1, self.stage2, self.stage3, self.stage4])
self.newly_added = nn.ModuleList([self.classifier])
self.init_params()
# self.loss = nn.MultiLabelSoftMarginLoss(weight=torch.tensor(cam_w, dtype=torch.float32))
self.loss = CB_loss(cam_w, n_classes, 'msl')
def __init__(self):
super(Net, self).__init__()
# backbone
self.resnet50 = resnet50.resnet50(pretrained=True, strides=[2, 2, 2, 1])
self.stage1 = nn.Sequential(self.resnet50.conv1, self.resnet50.bn1, self.resnet50.relu, self.resnet50.maxpool)
self.stage2 = nn.Sequential(self.resnet50.layer1)
self.stage3 = nn.Sequential(self.resnet50.layer2)
self.stage4 = nn.Sequential(self.resnet50.layer3)
self.stage5 = nn.Sequential(self.resnet50.layer4)
self.mean_shift = Net.MeanShift(2)
# branch: class boundary detection
self.fc_edge1 = nn.Sequential(
nn.Conv2d(64, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.ReLU(inplace=True),
)
self.fc_edge2 = nn.Sequential(
nn.Conv2d(256, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.ReLU(inplace=True),
)
self.fc_edge3 = nn.Sequential(
nn.Conv2d(512, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.Upsample(scale_factor=2, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_edge4 = nn.Sequential(
nn.Conv2d(1024, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.Upsample(scale_factor=4, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_edge5 = nn.Sequential(
nn.Conv2d(2048, 32, 1, bias=False),
nn.GroupNorm(4, 32),
nn.Upsample(scale_factor=4, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_edge6 = nn.Conv2d(160, 1, 1, bias=True)
# branch: displacement field
self.fc_dp1 = nn.Sequential(
nn.Conv2d(64, 64, 1, bias=False),
nn.GroupNorm(8, 64),
nn.ReLU(inplace=True),
)
self.fc_dp2 = nn.Sequential(
nn.Conv2d(256, 128, 1, bias=False),
nn.GroupNorm(16, 128),
nn.ReLU(inplace=True),
)
self.fc_dp3 = nn.Sequential(
nn.Conv2d(512, 256, 1, bias=False),
nn.GroupNorm(16, 256),
nn.ReLU(inplace=True),
)
self.fc_dp4 = nn.Sequential(
nn.Conv2d(1024, 256, 1, bias=False),
nn.GroupNorm(16, 256),
nn.Upsample(scale_factor=2, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_dp5 = nn.Sequential(
nn.Conv2d(2048, 256, 1, bias=False),
nn.GroupNorm(16, 256),
nn.Upsample(scale_factor=2, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_dp6 = nn.Sequential(
nn.Conv2d(768, 256, 1, bias=False),
nn.GroupNorm(16, 256),
nn.Upsample(scale_factor=2, mode='bilinear', align_corners=False),
nn.ReLU(inplace=True),
)
self.fc_dp7 = nn.Sequential(
nn.Conv2d(448, 256, 1, bias=False),
nn.GroupNorm(16, 256),
nn.ReLU(inplace=True),
nn.Conv2d(256, 2, 1, bias=False),
self.mean_shift
)
self.backbone = nn.ModuleList([self.stage1, self.stage2, self.stage3, self.stage4, self.stage5])
self.edge_layers = nn.ModuleList([self.fc_edge1, self.fc_edge2, self.fc_edge3, self.fc_edge4, self.fc_edge5, self.fc_edge6])
self.dp_layers = nn.ModuleList([self.fc_dp1, self.fc_dp2, self.fc_dp3, self.fc_dp4, self.fc_dp5, self.fc_dp6, self.fc_dp7])
