def __init__(self, **kwargs):
super(BoxNet, self).__init__()
self.conv_size = nn.Sequential(
resnet.resnet18(pretrained=False, input_channel=4, fc=False),
nn.Conv2d(512, 512, kernel_size=1, bias=False),
nn.BatchNorm2d(512), nn.ReLU(inplace=True), nn.AvgPool2d(7))
self.fc_size = nn.Sequential(nn.Linear(512, 256),
nn.ReLU(inplace=True), nn.Linear(256, 3))
self.conv_rot6 = nn.Sequential(
resnet.resnet18(pretrained=False, input_channel=4, fc=False),
nn.Conv2d(512, 512, kernel_size=1, bias=False),
nn.BatchNorm2d(512), nn.ReLU(inplace=True), nn.AvgPool2d(7))
self.fc_rot6 = nn.Sequential(nn.Linear(512, 256),
nn.ReLU(inplace=True), nn.Linear(256, 6),
nn.Tanh())
def __init__(self, **kwargs):
super(ScaleNet, self).__init__()
self.enc = resnet.resnet18(pretrained=False, input_channel=4, fc=False)
self.dec = nn.Sequential(
nn.Conv2d(1024, 512, kernel_size=1, bias=False),
nn.BatchNorm2d(512), nn.ReLU(inplace=True), nn.AvgPool2d(7))
self.fc = nn.Linear(512, 1)
def _build_model(self, inputs):
with tf.variable_scope('resnet'):
c2, c3, c4, c5 = resnet.resnet18(is_training=self.is_training).forward(inputs)
p5 = _conv(c5, 128, [1,1], is_training=self.is_training, name="conv_p5")
up_p5 = upsampling(p5)
reduce_dim_c4 = _conv(c4, 128, [1,1], is_training=self.is_training, name="conv_c4")
p4 = 0.5*up_p5 + 0.5*reduce_dim_c4
up_p4 = upsampling(p4)
reduce_dim_c3 = _conv(c3, 128, [1,1], is_training=self.is_training, name="conv_c3")
p3 = 0.5*up_p4 + 0.5*reduce_dim_c3
up_p3 = upsampling(p3)
reduce_dim_c2 = _conv(c2, 128, [1,1], is_training=self.is_training, name="conv_c2")
p2 = 0.5*up_p3 + 0.5*reduce_dim_c2
features = _conv(p2, 128, [3,3], is_training=self.is_training, name="conv_p2")
# IDA-up
# p2 = _conv(c2, 128, [1,1], is_training=self.is_training)
# p3 = _conv(c3, 128, [1,1], is_training=self.is_training)
# p4 = _conv(c4, 128, [1,1], is_training=self.is_training)
# p5 = _conv(c5, 128, [1,1], is_training=self.is_training)
# up_p3 = upsampling(p3, method='resize')
# p2 = _conv(p2+up_p3, 128, [3,3], is_training=self.is_training)
# up_p4 = upsampling(upsampling(p4, method='resize'), method='resize')
# p2 = _conv(p2+up_p4, 128, [3,3], is_training=self.is_training)
# up_p5 = upsampling(upsampling(upsampling(p5, method='resize'), method='resize'), method='resize')
# features = _conv(p2+up_p5, 128, [3,3], is_training=self.is_training)
with tf.variable_scope('detector'):
print('feature shape: ', features.shape)
hm = _conv(features, 64, [3,3], is_training=self.is_training, name="hm_conv_1")
hm = _conv_nn(hm, cfg.num_classes, [1, 1], padding='VALID', activation = tf.nn.relu, name='hm_conv')
#hm = _conv(hm, cfg.num_classes, [1, 1], padding="valid", name="hm")
wh = _conv(features, 64, [3,3], is_training=self.is_training, name="wh_conv_1")
#wh = tf.layers.conv2d(wh, 2, 1, 1, padding='valid', activation = None, bias_initializer=tf.constant_initializer(-np.log(99.)), name='wh')
wh = _conv_nn(wh, 2, [1, 1], padding='VALID', activation = tf.nn.relu, name="wh_conv")
#wh = tf.reshape(wh, [-1, wh.shape[1], wh.shape[2], wh.shape[3]])
#wh = tf.layers.conv2d(wh, 2, 1, 1, padding='valid', activation = None, name='wh')
#wh = _conv(wh, 2, [1, 1], padding="valid", name="wh")
reg = _conv(features, 64, [3,3], is_training=self.is_training, name="reg_conv_1")
#reg = tf.layers.conv2d(reg, 2, 1, 1, padding='valid', activation = None, bias_initializer=tf.constant_initializer(-np.log(99.)), name='reg')
reg = _conv_nn(reg, 2, [1, 1], padding='VALID', activation = tf.nn.relu, name="reg_conv")
#reg = tf.reshape(reg, [-1, reg.shape[1], reg.shape[2], reg.shape[3]])
#reg = _conv(reg, 2, [1, 1], padding="valid", name="reg")
return hm, wh, reg
def __init__(self, **kwargs):
super(BoxNet, self).__init__()
self.dec_size = nn.Sequential(
resnet.resnet18(pretrained=False,
input_channel=4,
fc=False,
norm=nn.InstanceNorm2d),
nn.Conv2d(512, 256, kernel_size=1, bias=False),
nn.InstanceNorm2d(256, affine=True), nn.ReLU(inplace=True),
nn.AvgPool2d(7), nn.Conv2d(256, 256, kernel_size=1, bias=True),
nn.ReLU(inplace=True), nn.Conv2d(256, 3, kernel_size=1, bias=True))
self.dec_rot6 = nn.Sequential(
resnet.resnet18(pretrained=False,
input_channel=4,
fc=False,
norm=nn.InstanceNorm2d),
nn.Conv2d(512, 256, kernel_size=1, bias=False),
nn.InstanceNorm2d(256, affine=True), nn.ReLU(inplace=True),
nn.AvgPool2d(7), nn.Conv2d(256, 256, kernel_size=1, bias=True),
nn.ReLU(inplace=True), nn.Conv2d(256, 6, kernel_size=1, bias=True),
nn.Tanh())
def __init__(self, **kwargs):
super(TouchPtNet, self).__init__()
self.enc = resnet.resnet18(pretrained=False, input_channel=4, fc=False)
self.dec = nn.Sequential(
nn.Conv2d(1024, 512, kernel_size=1, bias=False),
nn.BatchNorm2d(512), nn.ReLU(inplace=True), nn.AvgPool2d(7))
self.w1 = nn.Sequential(
nn.Conv1d(512 + 3, 256, kernel_size=1, bias=False),
nn.BatchNorm1d(256), nn.ReLU(inplace=True),
nn.Conv1d(256, 1, kernel_size=1, bias=True), nn.Softmax(dim=2))
self.w2 = nn.Sequential(
nn.Conv1d(512 + 3, 256, kernel_size=1, bias=False),
nn.BatchNorm1d(256), nn.ReLU(inplace=True),
nn.Conv1d(256, 1, kernel_size=1, bias=True), nn.Softmax(dim=2))