def initialize(self,
num_kpts,
temp_length,
input_dimension=3,
output_dimension=3,
output_pts=None):
self.num_kpts = num_kpts
self.temp_length = temp_length
self.output_dimension = output_dimension
self.output_pts = num_kpts if output_pts is None else output_pts
self.input_dimension = input_dimension
self.c1 = M.ConvLayer1D(3,
1024,
stride=3,
activation=M.PARAM_PRELU,
pad='VALID',
batch_norm=True,
usebias=False)
self.r1 = ResBlock1D(k=3)
self.r2 = ResBlock1D(k=3)
self.r3 = ResBlock1D(k=3)
self.r4 = ResBlock1D(k=3)
# self.r3 = ResBlock1D(k=3, dilation=3)
# self.c5 = M.ConvLayer1D(9, 256, activation=M.PARAM_PRELU, pad='VALID', batch_norm=True, usebias=False)
self.c4 = M.ConvLayer1D(1, self.output_pts * self.output_dimension)
def initialize(self, outchn=512, dilation=1, k=3):
self.bn = M.BatchNorm()
self.c1 = M.ConvLayer1D(k,
outchn,
dilation_rate=dilation,
activation=M.PARAM_PRELU,
batch_norm=True,
usebias=False,
pad='VALID')
self.c2 = M.ConvLayer1D(3, outchn, pad='VALID')
def initialize(self, num_kpts, temp_length): self.num_kpts = num_kpts self.temp_length = temp_length self.c1 = M.ConvLayer1D(3, 1024, stride=3, activation=M.PARAM_PRELU, pad='VALID', batch_norm=True, usebias=False) self.r1 = ResBlock1D(k=3) self.r2 = ResBlock1D(k=3) self.r3 = ResBlock1D(k=3) self.r4 = ResBlock1D(k=3) # self.r3 = ResBlock1D(k=3, dilation=3) # self.c5 = M.ConvLayer1D(9, 256, activation=M.PARAM_PRELU, pad='VALID', batch_norm=True, usebias=False) self.c4 = M.ConvLayer1D(1, num_kpts*3)
def initialize(self, outchn=1024, k=3):
self.k = k
# self.bn = M.BatchNorm()
self.c1 = M.ConvLayer1D(k,
outchn,
stride=k,
activation=M.PARAM_PRELU,
batch_norm=True,
usebias=False,
pad='VALID')
self.c2 = M.ConvLayer1D(1,
outchn,
activation=M.PARAM_PRELU,
batch_norm=True,
usebias=False,
pad='VALID')
def initialize(self):
self.c1 = M.ConvLayer1D(1,
1024,
activation=M.PARAM_PRELU,
batch_norm=True,
usebias=False)
self.c2 = M.ConvLayer1D(1,
256,
activation=M.PARAM_PRELU,
batch_norm=True,
usebias=False)
self.c3 = M.ConvLayer1D(1,
256,
activation=M.PARAM_PRELU,
batch_norm=True,
usebias=False)
self.c4 = M.ConvLayer1D(1, 1)
def initialize(self, num_pts, temp_length, pt_dim):
self.num_pts = num_pts
self.pt_dim = pt_dim
self.temp_length = temp_length
self.c1 = M.ConvLayer1D(5,
512,
activation=M.PARAM_PRELU,
batch_norm=True,
usebias=False,
pad='VALID')
self.r1 = ResBlock1D(k=3, dilation=2)
self.r2 = ResBlock1D(k=3, dilation=4)
self.r3 = ResBlock1D(k=5, dilation=8)
self.r4 = ResBlock1D(k=5, dilation=16)
self.c5 = M.ConvLayer1D(9,
512,
activation=M.PARAM_PRELU,
batch_norm=True,
usebias=False,
pad='VALID')
self.c4 = M.ConvLayer1D(1, num_pts * pt_dim)
def initialize(self):
self.c1 = M.ConvLayer1D(5,
512,
pad='VALID',
activation=M.PARAM_LRELU,
usebias=False,
batch_norm=True)
self.c2 = M.ConvLayer1D(5,
512,
dilation_rate=2,
pad='VALID',
activation=M.PARAM_LRELU,
usebias=False,
batch_norm=True)
self.c3 = M.ConvLayer1D(3,
512,
dilation_rate=4,
pad='VALID',
activation=M.PARAM_LRELU,
usebias=False,
batch_norm=True)
self.c4 = M.ConvLayer1D(3,
512,
dilation_rate=4,
pad='VALID',
activation=M.PARAM_LRELU,
usebias=False,
batch_norm=True)
self.c5 = M.ConvLayer1D(5,
512,
dilation_rate=8,
pad='VALID',
activation=M.PARAM_LRELU,
usebias=False,
batch_norm=True)
self.c6 = M.ConvLayer1D(5,
512,
pad='VALID',
activation=M.PARAM_LRELU,
usebias=False,
batch_norm=True)
self.c7 = M.ConvLayer1D(1, 62, usebias=False)
def initialize(self):
self.c1 = M.ConvLayer1D(1, 1024, activation=M.PARAM_PRELU)
self.c2 = M.ConvLayer1D(1, 256, activation=M.PARAM_PRELU)
self.c3 = M.ConvLayer1D(1, 256, activation=M.PARAM_PRELU)
self.c4 = M.ConvLayer1D(1, 1)
