Spaces:
Sleeping
Sleeping
import numpy as np | |
import torch | |
from torch import nn | |
import torch.nn.functional as F | |
from video3d.triplane_texture.ops import bias_act | |
from video3d.triplane_texture.ops import fma | |
from video3d.triplane_texture.ops import upfirdn2d | |
from video3d.triplane_texture.ops import conv2d_resample | |
from video3d.triplane_texture.ops import grid_sample_gradfix | |
from video3d.triplane_texture import misc | |
def modulated_conv2d( | |
x, # Input tensor of shape [batch_size, in_channels, in_height, in_width]. | |
weight, # Weight tensor of shape [out_channels, in_channels, kernel_height, kernel_width]. | |
styles, # Modulation coefficients of shape [batch_size, in_channels]. | |
noise=None, # Optional noise tensor to add to the output activations. | |
up=1, # Integer upsampling factor. | |
down=1, # Integer downsampling factor. | |
padding=0, # Padding with respect to the upsampled image. | |
resample_filter=None, | |
# Low-pass filter to apply when resampling activations. Must be prepared beforehand by calling upfirdn2d.setup_filter(). | |
demodulate=True, # Apply weight demodulation? | |
flip_weight=True, # False = convolution, True = correlation (matches torch.nn.functional.conv2d). | |
fused_modconv=True, # Perform modulation, convolution, and demodulation as a single fused operation? | |
): | |
batch_size = x.shape[0] | |
out_channels, in_channels, kh, kw = weight.shape | |
misc.assert_shape(weight, [out_channels, in_channels, kh, kw]) # [OIkk] | |
misc.assert_shape(x, [batch_size, in_channels, None, None]) # [NIHW] | |
misc.assert_shape(styles, [batch_size, in_channels]) # [NI] | |
# Pre-normalize inputs to avoid FP16 overflow. | |
if x.dtype == torch.float16 and demodulate: | |
weight = weight * (1 / np.sqrt(in_channels * kh * kw) / weight.norm( | |
float('inf'), dim=[1, 2, 3], keepdim=True)) # max_Ikk | |
styles = styles / styles.norm(float('inf'), dim=1, keepdim=True) # max_I | |
# Calculate per-sample weights and demodulation coefficients. | |
w = None | |
dcoefs = None | |
if demodulate or fused_modconv: | |
w = weight.unsqueeze(0) # [NOIkk] | |
w = w * styles.reshape(batch_size, 1, -1, 1, 1) # [NOIkk] | |
if demodulate: | |
dcoefs = (w.square().sum(dim=[2, 3, 4]) + 1e-8).rsqrt() # [NO] | |
if demodulate and fused_modconv: | |
w = w * dcoefs.reshape(batch_size, -1, 1, 1, 1) # [NOIkk] | |
# Execute by scaling the activations before and after the convolution. | |
if not fused_modconv: | |
x = x * styles.to(x.dtype).reshape(batch_size, -1, 1, 1) | |
x = conv2d_resample.conv2d_resample( | |
x=x, w=weight.to(x.dtype), f=resample_filter, up=up, down=down, padding=padding, flip_weight=flip_weight) | |
if demodulate and noise is not None: | |
x = fma.fma(x, dcoefs.to(x.dtype).reshape(batch_size, -1, 1, 1), noise.to(x.dtype)) | |
elif demodulate: | |
x = x * dcoefs.to(x.dtype).reshape(batch_size, -1, 1, 1) | |
elif noise is not None: | |
x = x.add_(noise.to(x.dtype)) | |
return x | |
# Execute as one fused op using grouped convolution. | |
with misc.suppress_tracer_warnings(): # this value will be treated as a constant | |
batch_size = int(batch_size) | |
misc.assert_shape(x, [batch_size, in_channels, None, None]) | |
x = x.reshape(1, -1, *x.shape[2:]) | |
w = w.reshape(-1, in_channels, kh, kw) | |
x = conv2d_resample.conv2d_resample( | |
x=x, w=w.to(x.dtype), f=resample_filter, up=up, down=down, padding=padding, groups=batch_size, | |
flip_weight=flip_weight) | |
x = x.reshape(batch_size, -1, *x.shape[2:]) | |
if noise is not None: | |
x = x.add_(noise) | |
return x | |
def modulated_fc( | |
x, # Input tensor of shape [batch_size, n_feature, in_channels]. | |
weight, # Weight tensor of shape [out_channels, in_channels]. | |
styles, # Modulation coefficients of shape [batch_size, in_channels]. | |
noise=None, # Optional noise tensor to add to the output activations. | |
demodulate=True, # Apply weight demodulation? | |
): | |
batch_size = x.shape[0] | |
n_feature = x.shape[1] | |
out_channels, in_channels = weight.shape | |
misc.assert_shape(weight, [out_channels, in_channels]) | |
misc.assert_shape(x, [batch_size, n_feature, in_channels]) | |
misc.assert_shape(styles, [batch_size, in_channels]) | |
assert demodulate | |
# Pre-normalize inputs to avoid FP16 overflow. | |
if x.dtype == torch.float16 and demodulate: | |
weight = weight * (1 / np.sqrt(in_channels) / weight.norm(float('inf'), dim=[1, 2, 3], keepdim=True)) # max_Ikk | |
styles = styles / styles.norm(float('inf'), dim=1, keepdim=True) # max_I | |
# Calculate per-sample weights and demodulation coefficients. | |
w = weight.unsqueeze(0) # [NOI] | |
w = w * styles.unsqueeze(dim=1) # [NOI] | |
dcoefs = (w.square().sum(dim=[2]) + 1e-8).rsqrt() # [NO] | |
w = w * dcoefs.unsqueeze(dim=-1) # [NOI] | |
x = torch.bmm(x, w.permute(0, 2, 1)) | |
if noise is not None: | |
x = x.add_(noise) | |
return x | |
class FullyConnectedLayer(torch.nn.Module): | |
def __init__( | |
self, | |
in_features, # Number of input features. | |
out_features, # Number of output features. | |
bias=True, # Apply additive bias before the activation function? | |
activation='linear', # Activation function: 'relu', 'lrelu', etc. | |
device='cuda', | |
lr_multiplier=1, # Learning rate multiplier. | |
bias_init=0, # Initial value for the additive bias. | |
): | |
super().__init__() | |
self.in_features = in_features | |
self.out_features = out_features | |
self.activation = activation | |
self.weight = torch.nn.Parameter(torch.randn([out_features, in_features], device=device) / lr_multiplier) | |
self.bias = torch.nn.Parameter( | |
torch.full([out_features], np.float32(bias_init), device=device)) if bias else None | |
self.weight_gain = lr_multiplier / np.sqrt(in_features) | |
self.bias_gain = lr_multiplier | |
def forward(self, x): | |
w = self.weight.to(x.dtype) * self.weight_gain | |
b = self.bias | |
if b is not None: | |
b = b.to(x.dtype) | |
if self.bias_gain != 1: | |
b = b * self.bias_gain | |
if self.activation == 'linear' and b is not None: | |
x = torch.addmm(b.unsqueeze(0), x, w.t()) | |
else: | |
x = x.matmul(w.t()) | |
x = bias_act.bias_act(x, b, act=self.activation) | |
return x | |
def extra_repr(self): | |
return f'in_features={self.in_features:d}, out_features={self.out_features:d}, activation={self.activation:s}' | |
class SynthesisLayer(torch.nn.Module): | |
def __init__( | |
self, | |
in_channels, # Number of input channels. | |
out_channels, # Number of output channels. | |
w_dim, # Intermediate latent (W) dimensionality. | |
resolution, # Resolution of this layer. | |
kernel_size=3, # Convolution kernel size. | |
up=1, # Integer upsampling factor. | |
use_noise=True, # Enable noise input? | |
activation='lrelu', # Activation function: 'relu', 'lrelu', etc. | |
device='cuda', | |
resample_filter=[1, 3, 3, 1], # Low-pass filter to apply when resampling activations. | |
conv_clamp=None, # Clamp the output of convolution layers to +-X, None = disable clamping. | |
channels_last=False, # Use channels_last format for the weights? | |
): | |
super().__init__() | |
self.in_channels = in_channels | |
self.out_channels = out_channels | |
self.w_dim = w_dim | |
self.resolution = resolution | |
self.up = up | |
self.use_noise = use_noise | |
self.activation = activation | |
self.conv_clamp = conv_clamp | |
self.register_buffer('resample_filter', upfirdn2d.setup_filter(resample_filter)) | |
self.padding = kernel_size // 2 | |
self.act_gain = bias_act.activation_funcs[activation].def_gain | |
self.affine = FullyConnectedLayer(w_dim, in_channels, bias_init=1, device=device) | |
memory_format = torch.channels_last if channels_last else torch.contiguous_format | |
self.weight = torch.nn.Parameter( | |
torch.randn([out_channels, in_channels, kernel_size, kernel_size], device=device).to( | |
memory_format=memory_format)) | |
if use_noise: | |
self.register_buffer('noise_const', torch.randn([resolution, resolution], device=device)) | |
self.noise_strength = torch.nn.Parameter(torch.zeros([], device=device)) | |
self.bias = torch.nn.Parameter(torch.zeros([out_channels], device=device)) | |
def forward(self, x, w, noise_mode='random', fused_modconv=True, gain=1): | |
assert noise_mode in ['random', 'const', 'none'] | |
in_resolution = self.resolution // self.up | |
misc.assert_shape(x, [None, self.in_channels, in_resolution, in_resolution]) | |
styles = self.affine(w) | |
noise = None | |
if self.use_noise and noise_mode == 'random': | |
noise = torch.randn( | |
[x.shape[0], 1, self.resolution, self.resolution], device=x.device) * self.noise_strength | |
if self.use_noise and noise_mode == 'const': | |
noise = self.noise_const * self.noise_strength | |
flip_weight = (self.up == 1) # slightly faster | |
x = modulated_conv2d( | |
x=x, weight=self.weight, styles=styles, noise=noise, up=self.up, | |
padding=self.padding, resample_filter=self.resample_filter, flip_weight=flip_weight, | |
fused_modconv=fused_modconv) | |
act_gain = self.act_gain * gain | |
act_clamp = self.conv_clamp * gain if self.conv_clamp is not None else None | |
x = bias_act.bias_act(x, self.bias.to(x.dtype), act=self.activation, gain=act_gain, clamp=act_clamp) | |
return x | |
def extra_repr(self): | |
return ' '.join( | |
[ | |
f'in_channels={self.in_channels:d}, out_channels={self.out_channels:d}, w_dim={self.w_dim:d},', | |
f'resolution={self.resolution:d}, up={self.up}, activation={self.activation:s}']) | |
class ImplicitSynthesisLayer(torch.nn.Module): | |
def __init__( | |
self, | |
in_channels, # Number of input channels. | |
out_channels, # Number of output channels. | |
w_dim, # Intermediate latent (W) dimensionality. | |
use_noise=True, # Enable noise input? | |
activation='lrelu', # Activation function: 'relu', 'lrelu', etc. | |
resample_filter=[1, 3, 3, 1], # Low-pass filter to apply when resampling activations. | |
device='cuda', | |
conv_clamp=None, # Clamp the output of convolution layers to +-X, None = disable clamping. | |
): | |
super().__init__() | |
self.in_channels = in_channels | |
self.out_channels = out_channels | |
self.w_dim = w_dim | |
self.use_noise = use_noise | |
self.activation = activation | |
self.conv_clamp = conv_clamp | |
self.act_gain = bias_act.activation_funcs[activation].def_gain | |
self.affine = FullyConnectedLayer(w_dim, in_channels, bias_init=1, device=device) | |
self.weight = torch.nn.Parameter(torch.randn([out_channels, in_channels], device=device)) | |
self.bias = torch.nn.Parameter(torch.zeros([out_channels], device=device)) | |
def forward(self, w, x, noise_mode='random', gain=1): | |
# x is the feature############# | |
# w is the condition | |
assert noise_mode in ['random', 'const', 'none'] | |
styles = self.affine(w) | |
noise = None # in te beegining, we didn't use the noise | |
x = modulated_fc(x=x, weight=self.weight, styles=styles, noise=noise) | |
act_gain = self.act_gain * gain | |
act_clamp = self.conv_clamp * gain if self.conv_clamp is not None else None | |
x = bias_act.bias_act( | |
x, self.bias.to(x.dtype), act=self.activation, gain=act_gain, clamp=act_clamp, | |
dim=2) # the last dim is the feature dim | |
return x | |
def extra_repr(self): | |
return ' '.join( | |
[ | |
f'in_channels={self.in_channels:d}, out_channels={self.out_channels:d}, w_dim={self.w_dim:d},', | |
f'activation={self.activation:s}']) | |
class Conv2dLayer(torch.nn.Module): | |
def __init__( | |
self, | |
in_channels, # Number of input channels. | |
out_channels, # Number of output channels. | |
kernel_size, # Width and height of the convolution kernel. | |
device='cuda', | |
bias=True, # Apply additive bias before the activation function? | |
activation='linear', # Activation function: 'relu', 'lrelu', etc. | |
up=1, # Integer upsampling factor. | |
down=1, # Integer downsampling factor. | |
resample_filter=[1, 3, 3, 1], # Low-pass filter to apply when resampling activations. | |
conv_clamp=None, # Clamp the output to +-X, None = disable clamping. | |
channels_last=False, # Expect the input to have memory_format=channels_last? | |
trainable=True, # Update the weights of this layer during training? | |
): | |
super().__init__() | |
self.in_channels = in_channels | |
self.out_channels = out_channels | |
self.activation = activation | |
self.up = up | |
self.down = down | |
self.conv_clamp = conv_clamp | |
self.register_buffer('resample_filter', upfirdn2d.setup_filter(resample_filter)) | |
self.padding = kernel_size // 2 | |
self.weight_gain = 1 / np.sqrt(in_channels * (kernel_size ** 2)) | |
self.act_gain = bias_act.activation_funcs[activation].def_gain | |
memory_format = torch.channels_last if channels_last else torch.contiguous_format | |
weight = torch.randn([out_channels, in_channels, kernel_size, kernel_size], device=device).to( | |
memory_format=memory_format) | |
bias = torch.zeros([out_channels], device=device) if bias else None | |
if trainable: | |
self.weight = torch.nn.Parameter(weight) | |
self.bias = torch.nn.Parameter(bias) if bias is not None else None | |
else: | |
self.register_buffer('weight', weight) | |
if bias is not None: | |
self.register_buffer('bias', bias) | |
else: | |
self.bias = None | |
def forward(self, x, gain=1): | |
w = self.weight * self.weight_gain | |
b = self.bias.to(x.dtype) if self.bias is not None else None | |
flip_weight = (self.up == 1) # slightly faster | |
x = conv2d_resample.conv2d_resample( | |
x=x, w=w.to(x.dtype), f=self.resample_filter, up=self.up, down=self.down, padding=self.padding, | |
flip_weight=flip_weight) | |
act_gain = self.act_gain * gain | |
act_clamp = self.conv_clamp * gain if self.conv_clamp is not None else None | |
x = bias_act.bias_act(x, b, act=self.activation, gain=act_gain, clamp=act_clamp) | |
return x | |
def extra_repr(self): | |
return ' '.join( | |
[ | |
f'in_channels={self.in_channels:d}, out_channels={self.out_channels:d}, activation={self.activation:s},', | |
f'up={self.up}, down={self.down}']) | |
class ToRGBLayer(torch.nn.Module): | |
def __init__( | |
self, in_channels, out_channels, w_dim, kernel_size=1, conv_clamp=None, channels_last=False, device='cuda'): | |
super().__init__() | |
self.in_channels = in_channels | |
self.out_channels = out_channels | |
self.w_dim = w_dim | |
self.conv_clamp = conv_clamp | |
self.affine = FullyConnectedLayer(w_dim, in_channels, bias_init=1, device=device) | |
memory_format = torch.channels_last if channels_last else torch.contiguous_format | |
self.weight = torch.nn.Parameter( | |
torch.randn([out_channels, in_channels, kernel_size, kernel_size], device=device).to( | |
memory_format=memory_format)) | |
self.bias = torch.nn.Parameter(torch.zeros([out_channels], device=device)) | |
self.weight_gain = 1 / np.sqrt(in_channels * (kernel_size ** 2)) | |
def forward(self, x, w, fused_modconv=True): | |
styles = self.affine(w) * self.weight_gain | |
x = modulated_conv2d(x=x, weight=self.weight, styles=styles, demodulate=False, fused_modconv=fused_modconv) | |
x = bias_act.bias_act(x, self.bias.to(x.dtype), clamp=self.conv_clamp) | |
return x | |
def extra_repr(self): | |
return f'in_channels={self.in_channels:d}, out_channels={self.out_channels:d}, w_dim={self.w_dim:d}' | |
class SynthesisBlock(torch.nn.Module): | |
def __init__( | |
self, | |
in_channels, # Number of input channels, 0 = first block. | |
out_channels, # Number of output channels. | |
w_dim, # Intermediate latent (W) dimensionality. | |
resolution, # Resolution of this block. | |
img_channels, # Number of output color channels. | |
is_last, # Is this the last block? | |
architecture='skip', # Architecture: 'orig', 'skip', 'resnet'. | |
resample_filter=[1, 3, 3, 1], # Low-pass filter to apply when resampling activations. | |
conv_clamp=256, # Clamp the output of convolution layers to +-X, None = disable clamping. | |
use_fp16=False, # Use FP16 for this block? | |
fp16_channels_last=False, # Use channels-last memory format with FP16? | |
fused_modconv_default=True, # Default value of fused_modconv. 'inference_only' = True for inference, False for training. | |
device='cuda', | |
first_layer=False, | |
**layer_kwargs, # Arguments for SynthesisLayer. | |
): | |
assert architecture in ['orig', 'skip', 'resnet'] | |
super().__init__() | |
self.first_layer = first_layer | |
self.in_channels = in_channels | |
self.w_dim = w_dim | |
self.resolution = resolution | |
self.img_channels = img_channels | |
self.is_last = is_last | |
self.architecture = architecture | |
self.use_fp16 = use_fp16 | |
self.channels_last = (use_fp16 and fp16_channels_last) | |
self.fused_modconv_default = fused_modconv_default | |
self.register_buffer('resample_filter', upfirdn2d.setup_filter(resample_filter)) | |
self.num_conv = 0 | |
self.num_torgb = 0 | |
if in_channels == 0: | |
self.const = torch.nn.Parameter(torch.randn([out_channels, resolution, resolution], device=device)) | |
if in_channels != 0: | |
if self.first_layer: | |
self.conv0 = SynthesisLayer( | |
in_channels, out_channels, w_dim=w_dim, resolution=resolution, | |
conv_clamp=conv_clamp, channels_last=self.channels_last, device=device, **layer_kwargs) | |
else: | |
self.conv0 = SynthesisLayer( | |
in_channels, out_channels, w_dim=w_dim, resolution=resolution, up=2, | |
resample_filter=resample_filter, conv_clamp=conv_clamp, channels_last=self.channels_last, device=device, | |
**layer_kwargs) | |
self.num_conv += 1 | |
self.conv1 = SynthesisLayer( | |
out_channels, out_channels, w_dim=w_dim, resolution=resolution, | |
conv_clamp=conv_clamp, channels_last=self.channels_last, device=device, **layer_kwargs) | |
self.num_conv += 1 | |
if is_last or architecture == 'skip': | |
self.torgb = ToRGBLayer( | |
out_channels, img_channels, w_dim=w_dim, | |
conv_clamp=conv_clamp, channels_last=self.channels_last, device=device) | |
self.num_torgb += 1 | |
if in_channels != 0 and architecture == 'resnet': | |
self.skip = Conv2dLayer( | |
in_channels, out_channels, kernel_size=1, bias=False, up=2, | |
resample_filter=resample_filter, channels_last=self.channels_last, device=device) | |
def forward(self, x, img, ws, force_fp32=False, fused_modconv='inference_only', update_emas=False, **layer_kwargs): | |
_ = update_emas # unused | |
misc.assert_shape(ws, [None, self.num_conv + self.num_torgb, self.w_dim]) | |
w_iter = iter(ws.unbind(dim=1)) | |
if ws.device.type != 'cuda': | |
force_fp32 = True | |
dtype = torch.float16 if self.use_fp16 and not force_fp32 else torch.float32 | |
memory_format = torch.channels_last if self.channels_last and not force_fp32 else torch.contiguous_format | |
if fused_modconv is None: | |
fused_modconv = self.fused_modconv_default | |
if fused_modconv == 'inference_only': | |
fused_modconv = (not self.training) | |
## | |
# Input. | |
if self.in_channels == 0: | |
x = self.const.to(dtype=dtype, memory_format=memory_format) | |
x = x.unsqueeze(0).repeat([ws.shape[0], 1, 1, 1]) | |
else: | |
# misc.assert_shape(x, [None, self.in_channels, self.resolution // 2, self.resolution // 2]) | |
x = x.to(dtype=dtype, memory_format=memory_format) | |
# Main layers. | |
if self.in_channels == 0: | |
x = self.conv1(x, next(w_iter), fused_modconv=fused_modconv, **layer_kwargs) | |
elif self.architecture == 'resnet': | |
y = self.skip(x, gain=np.sqrt(0.5)) | |
x = self.conv0(x, next(w_iter), fused_modconv=fused_modconv, **layer_kwargs) | |
x = self.conv1(x, next(w_iter), fused_modconv=fused_modconv, gain=np.sqrt(0.5), **layer_kwargs) | |
x = y + x | |
else: | |
x = self.conv0(x, next(w_iter), fused_modconv=fused_modconv, **layer_kwargs) | |
x = self.conv1(x, next(w_iter), fused_modconv=fused_modconv, **layer_kwargs) | |
if img is not None: | |
misc.assert_shape(img, [None, self.img_channels, self.resolution // 2, self.resolution // 2]) | |
img = upfirdn2d.upsample2d(img, self.resample_filter) | |
if self.is_last or self.architecture == 'skip': | |
y = self.torgb(x, next(w_iter), fused_modconv=fused_modconv) | |
y = y.to(dtype=torch.float32, memory_format=torch.contiguous_format) | |
img = img + y if img is not None else y | |
assert x.dtype == dtype | |
assert img is None or img.dtype == torch.float32 | |
return x, img | |
def extra_repr(self): | |
return f'resolution={self.resolution:d}, architecture={self.architecture:s}' | |
class SynthesisNetwork(torch.nn.Module): | |
def __init__( | |
self, | |
w_dim, # Intermediate latent (W) dimensionality. | |
img_resolution, # Output image resolution. | |
img_channels, # Number of color channels. | |
channel_base=32768, # Overall multiplier for the number of channels. | |
channel_max=512, # Maximum number of channels in any layer. | |
num_fp16_res=4, # Use FP16 for the N highest resolutions. | |
device='cuda', | |
**block_kwargs, # Arguments for SynthesisBlock. | |
): | |
assert img_resolution >= 4 and img_resolution & (img_resolution - 1) == 0 | |
super().__init__() | |
self.w_dim = w_dim | |
self.img_resolution = img_resolution | |
self.img_resolution_log2 = int(np.log2(img_resolution)) | |
self.img_channels = img_channels | |
self.num_fp16_res = num_fp16_res | |
self.block_resolutions = [2 ** i for i in range(2, self.img_resolution_log2 + 1)] # [4,8,16,32,64,128] | |
# {4: 512, 8: 512, 16: 512, 32: 512, 64: 512, 128: 256} | |
channels_dict = {res: min(channel_base // res, channel_max) for res in self.block_resolutions} | |
self.num_ws = 0 | |
for res in self.block_resolutions: | |
in_channels = channels_dict[res // 2] if res > 4 else 0 | |
out_channels = channels_dict[res] | |
is_last = (res == self.img_resolution) | |
use_fp16 = False | |
block = SynthesisBlock( | |
in_channels, out_channels, w_dim=w_dim, resolution=res, | |
img_channels=img_channels, is_last=is_last, use_fp16=use_fp16, device=device, **block_kwargs) | |
self.num_ws += block.num_conv | |
self.num_ws += block.num_torgb | |
setattr(self, f'b{res}', block) | |
def forward(self, ws, **block_kwargs): | |
block_ws = [] | |
misc.assert_shape(ws, [None, self.num_ws, self.w_dim]) | |
ws = ws.to(torch.float32) | |
w_idx = 0 | |
for res in self.block_resolutions: | |
block = getattr(self, f'b{res}') | |
block_ws.append(ws.narrow(1, w_idx, block.num_conv + block.num_torgb)) | |
w_idx += (block.num_conv + block.num_torgb) | |
x = img = None | |
for res, cur_ws in zip(self.block_resolutions, block_ws): | |
block = getattr(self, f'b{res}') | |
x, img = block(x, img, cur_ws, **block_kwargs) | |
return img | |
def extra_repr(self): | |
return ' '.join( | |
[ | |
f'w_dim={self.w_dim:d}, num_ws={self.num_ws:d},', | |
f'img_resolution={self.img_resolution:d}, img_channels={self.img_channels:d},', | |
f'num_fp16_res={self.num_fp16_res:d}']) | |
class ImplicitSynthesisNetwork(torch.nn.Module): | |
def __init__( | |
self, | |
w_dim=512, # Intermediate latent (W) dimensionality. | |
input_channel=256, | |
out_channels=3, # Number of color channels. | |
latent_channel=256, | |
n_layers=4, | |
device='cuda' | |
): | |
super().__init__() | |
self.n_layer = n_layers | |
self.layers = [] | |
self.num_ws = 0 | |
for i_layer in range(self.n_layer): | |
layer = ImplicitSynthesisLayer( | |
w_dim=w_dim, | |
in_channels=input_channel if i_layer == 0 else latent_channel, | |
out_channels=latent_channel, device=device) | |
self.layers.append(layer) | |
self.num_ws += 1 | |
self.layers.append( | |
ImplicitSynthesisLayer( | |
w_dim=w_dim, in_channels=latent_channel, out_channels=out_channels, | |
activation='sigmoid', device=device) | |
) | |
self.num_ws += 1 | |
self.layers = torch.nn.ModuleList(self.layers) | |
self.w_dim = w_dim | |
self.out_channels = out_channels | |
def forward(self, ws, position, **block_kwargs): | |
out = position | |
for i in range(self.n_layer): | |
out = self.layers[i](ws[:, i], out) | |
out = self.layers[-1](ws[:, self.n_layer], out) | |
return out | |
def extra_repr(self): | |
return ' '.join( | |
[ | |
f'w_dim={self.w_dim:d}']) | |
class TriPlaneTex(torch.nn.Module): | |
def __init__( | |
self, | |
w_dim, # Intermediate latent (W) dimensionality. | |
img_channels, # Number of color channels. | |
tri_plane_resolution=256, | |
device='cuda', | |
mlp_latent_channel=256, | |
n_implicit_layer=3, | |
feat_dim=384, # number of feat dim from encoder | |
n_mapping_layer=8, | |
sym_texture=True, | |
grid_scale=7., | |
min_max=None, | |
perturb_normal=False, | |
**block_kwargs, # Arguments for SynthesisBlock. | |
): | |
super().__init__() | |
self.n_implicit_layer = n_implicit_layer | |
self.img_feat_dim = 32 # The setting follows Koki's paper | |
self.w_dim = w_dim | |
self.tri_plane_resolution = tri_plane_resolution | |
# the mapping network | |
self.feat_dim = feat_dim | |
self.n_mapping_layer = n_mapping_layer | |
self.embed = FullyConnectedLayer(feat_dim, w_dim, device=device) | |
for idx in range(n_mapping_layer): | |
layer = FullyConnectedLayer(w_dim, w_dim, activation='lrelu', lr_multiplier=0.1, device=device) | |
setattr(self, f'mapping{idx}', layer) | |
# self.w_dim = w_dim * 2 | |
self.tri_plane_synthesis = SynthesisNetwork( | |
w_dim=self.w_dim, img_resolution=self.tri_plane_resolution, | |
img_channels=self.img_feat_dim * 3, | |
device=device, | |
**block_kwargs) | |
self.num_ws_tri_plane = self.tri_plane_synthesis.num_ws | |
mlp_input_channel = self.img_feat_dim + w_dim # | |
mlp_latent_channel = mlp_latent_channel | |
mlp_input_channel -= w_dim | |
self.mlp_synthesis = ImplicitSynthesisNetwork( | |
out_channels=img_channels, | |
n_layers=self.n_implicit_layer, | |
w_dim=self.w_dim, | |
latent_channel=mlp_latent_channel, | |
input_channel=mlp_input_channel, | |
device=device) | |
self.num_ws_all = self.num_ws_tri_plane + self.mlp_synthesis.num_ws | |
# texture related | |
self.sym_texture = sym_texture | |
self.grid_scale = grid_scale | |
self.shape_min = 0. | |
self.shape_lenght = grid_scale / 2. | |
if min_max is not None: | |
self.register_buffer('min_max', min_max) | |
else: | |
self.min_max = None | |
self.perturb_normal = perturb_normal | |
def old_forward( | |
self, feat, position=None, **block_kwargs): | |
''' | |
Predict texture with given latent code | |
:param feat: image global feat | |
:param position: position for the surface points | |
:param block_kwargs: | |
:return: | |
''' | |
assert feat.shape[-1] == self.feat_dim | |
# mapping global feature to ws | |
ws = self.embed(feat) | |
for idx in range(self.n_mapping_layer): | |
layer = getattr(self, f'mapping{idx}') | |
ws = layer(ws) | |
ws = ws.unsqueeze(1).repeat(1, self.num_ws_all, 1) | |
plane_feat = self.tri_plane_synthesis(ws[:, :self.num_ws_tri_plane], **block_kwargs) | |
tri_plane = torch.split(plane_feat, self.img_feat_dim, dim=1) | |
normalized_tex_pos = (position - self.shape_min) / self.shape_lenght # in [-1, 1] | |
normalized_tex_pos = torch.clamp(normalized_tex_pos, -1.0, 1.0) | |
if self.sym_texture: | |
x_pos, y_pos, z_pos = normalized_tex_pos.unbind(-1) | |
normalized_tex_pos = torch.stack([x_pos.abs(), y_pos, z_pos], dim=-1) | |
x_feat = grid_sample_gradfix.grid_sample( | |
tri_plane[0], | |
torch.cat( | |
[normalized_tex_pos[:, :, 0:1], normalized_tex_pos[:, :, 1:2]], | |
dim=-1).unsqueeze(dim=1).detach()) | |
y_feat = grid_sample_gradfix.grid_sample( | |
tri_plane[1], | |
torch.cat( | |
[normalized_tex_pos[:, :, 1:2], normalized_tex_pos[:, :, 2:3]], | |
dim=-1).unsqueeze(dim=1).detach()) | |
z_feat = grid_sample_gradfix.grid_sample( | |
tri_plane[2], | |
torch.cat( | |
[normalized_tex_pos[:, :, 0:1], normalized_tex_pos[:, :, 2:3]], | |
dim=-1).unsqueeze(dim=1).detach()) | |
final_feat = (x_feat + y_feat + z_feat) | |
final_feat = final_feat.squeeze(dim=2).permute(0, 2, 1) # 32dimension | |
final_feat_tex = final_feat | |
out = self.mlp_synthesis(ws[:, self.num_ws_tri_plane:], final_feat_tex) | |
return out | |
def sample(self, xyz, feat=None, feat_map=None, mvp=None, w2c=None, deform_xyz=None): | |
# query the deformed points or canonical points | |
# x = deform_xyz | |
x = xyz | |
b, h, w, c = x.shape | |
mvp = mvp.detach() # [b, 4, 4] | |
w2c = w2c.detach() # [b, 4, 4] | |
x = x.reshape(b, -1, c) | |
global_feat = feat # [b, d] | |
out = self.old_forward( | |
feat=global_feat, | |
position=x | |
) | |
if self.min_max is not None: | |
out = out * (self.min_max[1][None, :] - self.min_max[0][None, :]) + self.min_max[0][None, :] | |
return out.view(b, h, w, -1) |