File size: 10,548 Bytes
f12ab4c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 |
import torch
import torch.nn as nn
import torch.nn.functional as F
import torchvision
from resnet import Resnet18
# from modules.bn import InPlaceABNSync as BatchNorm2d
class ConvBNReLU(nn.Module):
def __init__(self, in_chan, out_chan, ks=3, stride=1, padding=1, *args, **kwargs):
super(ConvBNReLU, self).__init__()
self.conv = nn.Conv2d(in_chan,
out_chan,
kernel_size = ks,
stride = stride,
padding = padding,
bias = False)
self.bn = nn.BatchNorm2d(out_chan)
self.init_weight()
def forward(self, x):
x = self.conv(x)
x = F.relu(self.bn(x))
return x
def init_weight(self):
for ly in self.children():
if isinstance(ly, nn.Conv2d):
nn.init.kaiming_normal_(ly.weight, a=1)
if not ly.bias is None: nn.init.constant_(ly.bias, 0)
class BiSeNetOutput(nn.Module):
def __init__(self, in_chan, mid_chan, n_classes, *args, **kwargs):
super(BiSeNetOutput, self).__init__()
self.conv = ConvBNReLU(in_chan, mid_chan, ks=3, stride=1, padding=1)
self.conv_out = nn.Conv2d(mid_chan, n_classes, kernel_size=1, bias=False)
self.init_weight()
def forward(self, x):
x = self.conv(x)
x = self.conv_out(x)
return x
def init_weight(self):
for ly in self.children():
if isinstance(ly, nn.Conv2d):
nn.init.kaiming_normal_(ly.weight, a=1)
if not ly.bias is None: nn.init.constant_(ly.bias, 0)
def get_params(self):
wd_params, nowd_params = [], []
for name, module in self.named_modules():
if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d):
wd_params.append(module.weight)
if not module.bias is None:
nowd_params.append(module.bias)
elif isinstance(module, nn.BatchNorm2d):
nowd_params += list(module.parameters())
return wd_params, nowd_params
class AttentionRefinementModule(nn.Module):
def __init__(self, in_chan, out_chan, *args, **kwargs):
super(AttentionRefinementModule, self).__init__()
self.conv = ConvBNReLU(in_chan, out_chan, ks=3, stride=1, padding=1)
self.conv_atten = nn.Conv2d(out_chan, out_chan, kernel_size= 1, bias=False)
self.bn_atten = nn.BatchNorm2d(out_chan)
self.sigmoid_atten = nn.Sigmoid()
self.init_weight()
def forward(self, x):
feat = self.conv(x)
atten = F.avg_pool2d(feat, feat.size()[2:])
atten = self.conv_atten(atten)
atten = self.bn_atten(atten)
atten = self.sigmoid_atten(atten)
out = torch.mul(feat, atten)
return out
def init_weight(self):
for ly in self.children():
if isinstance(ly, nn.Conv2d):
nn.init.kaiming_normal_(ly.weight, a=1)
if not ly.bias is None: nn.init.constant_(ly.bias, 0)
class ContextPath(nn.Module):
def __init__(self, *args, **kwargs):
super(ContextPath, self).__init__()
self.resnet = Resnet18()
self.arm16 = AttentionRefinementModule(256, 128)
self.arm32 = AttentionRefinementModule(512, 128)
self.conv_head32 = ConvBNReLU(128, 128, ks=3, stride=1, padding=1)
self.conv_head16 = ConvBNReLU(128, 128, ks=3, stride=1, padding=1)
self.conv_avg = ConvBNReLU(512, 128, ks=1, stride=1, padding=0)
self.init_weight()
def forward(self, x):
H0, W0 = x.size()[2:]
feat8, feat16, feat32 = self.resnet(x)
H8, W8 = feat8.size()[2:]
H16, W16 = feat16.size()[2:]
H32, W32 = feat32.size()[2:]
avg = F.avg_pool2d(feat32, feat32.size()[2:])
avg = self.conv_avg(avg)
avg_up = F.interpolate(avg, (H32, W32), mode='nearest')
feat32_arm = self.arm32(feat32)
feat32_sum = feat32_arm + avg_up
feat32_up = F.interpolate(feat32_sum, (H16, W16), mode='nearest')
feat32_up = self.conv_head32(feat32_up)
feat16_arm = self.arm16(feat16)
feat16_sum = feat16_arm + feat32_up
feat16_up = F.interpolate(feat16_sum, (H8, W8), mode='nearest')
feat16_up = self.conv_head16(feat16_up)
return feat8, feat16_up, feat32_up # x8, x8, x16
def init_weight(self):
for ly in self.children():
if isinstance(ly, nn.Conv2d):
nn.init.kaiming_normal_(ly.weight, a=1)
if not ly.bias is None: nn.init.constant_(ly.bias, 0)
def get_params(self):
wd_params, nowd_params = [], []
for name, module in self.named_modules():
if isinstance(module, (nn.Linear, nn.Conv2d)):
wd_params.append(module.weight)
if not module.bias is None:
nowd_params.append(module.bias)
elif isinstance(module, nn.BatchNorm2d):
nowd_params += list(module.parameters())
return wd_params, nowd_params
### This is not used, since I replace this with the resnet feature with the same size
class SpatialPath(nn.Module):
def __init__(self, *args, **kwargs):
super(SpatialPath, self).__init__()
self.conv1 = ConvBNReLU(3, 64, ks=7, stride=2, padding=3)
self.conv2 = ConvBNReLU(64, 64, ks=3, stride=2, padding=1)
self.conv3 = ConvBNReLU(64, 64, ks=3, stride=2, padding=1)
self.conv_out = ConvBNReLU(64, 128, ks=1, stride=1, padding=0)
self.init_weight()
def forward(self, x):
feat = self.conv1(x)
feat = self.conv2(feat)
feat = self.conv3(feat)
feat = self.conv_out(feat)
return feat
def init_weight(self):
for ly in self.children():
if isinstance(ly, nn.Conv2d):
nn.init.kaiming_normal_(ly.weight, a=1)
if not ly.bias is None: nn.init.constant_(ly.bias, 0)
def get_params(self):
wd_params, nowd_params = [], []
for name, module in self.named_modules():
if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d):
wd_params.append(module.weight)
if not module.bias is None:
nowd_params.append(module.bias)
elif isinstance(module, nn.BatchNorm2d):
nowd_params += list(module.parameters())
return wd_params, nowd_params
class FeatureFusionModule(nn.Module):
def __init__(self, in_chan, out_chan, *args, **kwargs):
super(FeatureFusionModule, self).__init__()
self.convblk = ConvBNReLU(in_chan, out_chan, ks=1, stride=1, padding=0)
self.conv1 = nn.Conv2d(out_chan,
out_chan//4,
kernel_size = 1,
stride = 1,
padding = 0,
bias = False)
self.conv2 = nn.Conv2d(out_chan//4,
out_chan,
kernel_size = 1,
stride = 1,
padding = 0,
bias = False)
self.relu = nn.ReLU(inplace=True)
self.sigmoid = nn.Sigmoid()
self.init_weight()
def forward(self, fsp, fcp):
fcat = torch.cat([fsp, fcp], dim=1)
feat = self.convblk(fcat)
atten = F.avg_pool2d(feat, feat.size()[2:])
atten = self.conv1(atten)
atten = self.relu(atten)
atten = self.conv2(atten)
atten = self.sigmoid(atten)
feat_atten = torch.mul(feat, atten)
feat_out = feat_atten + feat
return feat_out
def init_weight(self):
for ly in self.children():
if isinstance(ly, nn.Conv2d):
nn.init.kaiming_normal_(ly.weight, a=1)
if not ly.bias is None: nn.init.constant_(ly.bias, 0)
def get_params(self):
wd_params, nowd_params = [], []
for name, module in self.named_modules():
if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d):
wd_params.append(module.weight)
if not module.bias is None:
nowd_params.append(module.bias)
elif isinstance(module, nn.BatchNorm2d):
nowd_params += list(module.parameters())
return wd_params, nowd_params
class BiSeNet(nn.Module):
def __init__(self, n_classes, *args, **kwargs):
super(BiSeNet, self).__init__()
self.cp = ContextPath()
## here self.sp is deleted
self.ffm = FeatureFusionModule(256, 256)
self.conv_out = BiSeNetOutput(256, 256, n_classes)
self.conv_out16 = BiSeNetOutput(128, 64, n_classes)
self.conv_out32 = BiSeNetOutput(128, 64, n_classes)
self.init_weight()
def forward(self, x):
H, W = x.size()[2:]
feat_res8, feat_cp8, feat_cp16 = self.cp(x) # here return res3b1 feature
feat_sp = feat_res8 # use res3b1 feature to replace spatial path feature
feat_fuse = self.ffm(feat_sp, feat_cp8)
feat_out = self.conv_out(feat_fuse)
feat_out16 = self.conv_out16(feat_cp8)
feat_out32 = self.conv_out32(feat_cp16)
feat_out = F.interpolate(feat_out, (H, W), mode='bilinear', align_corners=True)
feat_out16 = F.interpolate(feat_out16, (H, W), mode='bilinear', align_corners=True)
feat_out32 = F.interpolate(feat_out32, (H, W), mode='bilinear', align_corners=True)
return feat_out, feat_out16, feat_out32
def init_weight(self):
for ly in self.children():
if isinstance(ly, nn.Conv2d):
nn.init.kaiming_normal_(ly.weight, a=1)
if not ly.bias is None: nn.init.constant_(ly.bias, 0)
def get_params(self):
wd_params, nowd_params, lr_mul_wd_params, lr_mul_nowd_params = [], [], [], []
for name, child in self.named_children():
child_wd_params, child_nowd_params = child.get_params()
if isinstance(child, FeatureFusionModule) or isinstance(child, BiSeNetOutput):
lr_mul_wd_params += child_wd_params
lr_mul_nowd_params += child_nowd_params
else:
wd_params += child_wd_params
nowd_params += child_nowd_params
return wd_params, nowd_params, lr_mul_wd_params, lr_mul_nowd_params
if __name__ == "__main__":
net = BiSeNet(19)
net.cuda()
net.eval()
in_ten = torch.randn(16, 3, 640, 480).cuda()
out, out16, out32 = net(in_ten)
print(out.shape)
net.get_params()
|