|
|
import torch
|
|
|
from torch import nn
|
|
|
from torch.nn import functional as F
|
|
|
import pdb
|
|
|
from .conv import Conv2dTranspose, Conv2d, nonorm_Conv2d
|
|
|
|
|
|
|
|
|
class Wav2Lip(nn.Module):
|
|
|
def __init__(self):
|
|
|
super(Wav2Lip, self).__init__()
|
|
|
|
|
|
self.face_encoder_blocks = nn.ModuleList([
|
|
|
nn.Sequential(Conv2d(6, 16, kernel_size=7, stride=1, padding=3)),
|
|
|
|
|
|
nn.Sequential(Conv2d(16, 32, kernel_size=3, stride=2, padding=1),
|
|
|
Conv2d(32, 32, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(32, 32, kernel_size=3, stride=1, padding=1, residual=True)),
|
|
|
|
|
|
nn.Sequential(Conv2d(32, 64, kernel_size=3, stride=2, padding=1),
|
|
|
Conv2d(64, 64, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(64, 64, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(64, 64, kernel_size=3, stride=1, padding=1, residual=True)),
|
|
|
|
|
|
nn.Sequential(Conv2d(64, 128, kernel_size=3, stride=2, padding=1),
|
|
|
Conv2d(128, 128, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(128, 128, kernel_size=3, stride=1, padding=1, residual=True)),
|
|
|
|
|
|
nn.Sequential(Conv2d(128, 256, kernel_size=3, stride=2, padding=1),
|
|
|
Conv2d(256, 256, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(256, 256, kernel_size=3, stride=1, padding=1, residual=True)),
|
|
|
|
|
|
nn.Sequential(Conv2d(256, 512, kernel_size=3, stride=2, padding=1),
|
|
|
Conv2d(512, 512, kernel_size=3, stride=1, padding=1, residual=True), ),
|
|
|
|
|
|
nn.Sequential(Conv2d(512, 512, kernel_size=3, stride=2, padding=1),
|
|
|
Conv2d(512, 512, kernel_size=3, stride=1, padding=1, residual=True), ),
|
|
|
|
|
|
nn.Sequential(Conv2d(512, 512, kernel_size=4, stride=1, padding=0),
|
|
|
Conv2d(512, 512, kernel_size=1, stride=1, padding=0)), ])
|
|
|
|
|
|
self.audio_encoder = nn.Sequential(
|
|
|
Conv2d(1, 32, kernel_size=3, stride=1, padding=1),
|
|
|
Conv2d(32, 32, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(32, 32, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
|
|
|
Conv2d(32, 64, kernel_size=3, stride=(3, 1), padding=1),
|
|
|
Conv2d(64, 64, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(64, 64, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
|
|
|
Conv2d(64, 128, kernel_size=3, stride=3, padding=1),
|
|
|
Conv2d(128, 128, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(128, 128, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
|
|
|
Conv2d(128, 256, kernel_size=3, stride=(3, 2), padding=1),
|
|
|
Conv2d(256, 256, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
|
|
|
Conv2d(256, 512, kernel_size=3, stride=1, padding=0),
|
|
|
Conv2d(512, 512, kernel_size=1, stride=1, padding=0), )
|
|
|
|
|
|
self.face_decoder_blocks = nn.ModuleList([
|
|
|
nn.Sequential(Conv2d(512, 512, kernel_size=1, stride=1, padding=0), ),
|
|
|
|
|
|
nn.Sequential(Conv2dTranspose(1024, 512, kernel_size=4, stride=1, padding=0),
|
|
|
Conv2d(512, 512, kernel_size=3, stride=1, padding=1, residual=True), ),
|
|
|
|
|
|
nn.Sequential(Conv2dTranspose(1024, 512, kernel_size=3, stride=2, padding=1, output_padding=1),
|
|
|
Conv2d(512, 512, kernel_size=3, stride=1, padding=1, residual=True), ),
|
|
|
|
|
|
nn.Sequential(Conv2dTranspose(1024, 512, kernel_size=3, stride=2, padding=1, output_padding=1),
|
|
|
Conv2d(512, 512, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(512, 512, kernel_size=3, stride=1, padding=1, residual=True), ),
|
|
|
|
|
|
nn.Sequential(Conv2dTranspose(768, 384, kernel_size=3, stride=2, padding=1, output_padding=1),
|
|
|
Conv2d(384, 384, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(384, 384, kernel_size=3, stride=1, padding=1, residual=True), ),
|
|
|
|
|
|
nn.Sequential(Conv2dTranspose(512, 256, kernel_size=3, stride=2, padding=1, output_padding=1),
|
|
|
Conv2d(256, 256, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(256, 256, kernel_size=3, stride=1, padding=1, residual=True), ),
|
|
|
|
|
|
nn.Sequential(Conv2dTranspose(320, 128, kernel_size=3, stride=2, padding=1, output_padding=1),
|
|
|
Conv2d(128, 128, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(128, 128, kernel_size=3, stride=1, padding=1, residual=True), ),
|
|
|
|
|
|
nn.Sequential(Conv2dTranspose(160, 64, kernel_size=3, stride=2, padding=1, output_padding=1),
|
|
|
Conv2d(64, 64, kernel_size=3, stride=1, padding=1, residual=True),
|
|
|
Conv2d(64, 64, kernel_size=3, stride=1, padding=1, residual=True), ), ])
|
|
|
|
|
|
self.output_block = nn.Sequential(Conv2d(80, 32, kernel_size=3, stride=1, padding=1),
|
|
|
nn.Conv2d(32, 3, kernel_size=1, stride=1, padding=0),
|
|
|
nn.Sigmoid())
|
|
|
|
|
|
def audio_forward(self, audio_sequences, a_alpha=1.):
|
|
|
audio_embedding = self.audio_encoder(audio_sequences) # B, 512, 1, 1
|
|
|
if a_alpha != 1.:
|
|
|
audio_embedding *= a_alpha
|
|
|
return audio_embedding
|
|
|
|
|
|
def inference(self, audio_embedding, face_sequences):
|
|
|
feats = []
|
|
|
x = face_sequences
|
|
|
for f in self.face_encoder_blocks:
|
|
|
x = f(x)
|
|
|
feats.append(x)
|
|
|
|
|
|
x = audio_embedding
|
|
|
for f in self.face_decoder_blocks:
|
|
|
x = f(x)
|
|
|
try:
|
|
|
x = torch.cat((x, feats[-1]), dim=1)
|
|
|
except Exception as e:
|
|
|
print(x.size())
|
|
|
print(feats[-1].size())
|
|
|
raise e
|
|
|
|
|
|
feats.pop()
|
|
|
|
|
|
x = self.output_block(x)
|
|
|
outputs = x
|
|
|
|
|
|
return outputs
|
|
|
|
|
|
def forward(self, audio_sequences, face_sequences, a_alpha=1.):
|
|
|
# audio_sequences = (B, T, 1, 80, 16)
|
|
|
B = audio_sequences.size(0)
|
|
|
|
|
|
input_dim_size = len(face_sequences.size())
|
|
|
if input_dim_size > 4:
|
|
|
audio_sequences = torch.cat([audio_sequences[:, i] for i in range(audio_sequences.size(1))], dim=0)#[bz, 5, 1, 80, 16]->[bz*5, 1, 80, 16]
|
|
|
face_sequences = torch.cat([face_sequences[:, :, i] for i in range(face_sequences.size(2))], dim=0)#[bz, 6, 5, 256, 256]->[bz*5, 6, 256, 256]
|
|
|
|
|
|
audio_embedding = self.audio_encoder(audio_sequences) # [bz*5, 1, 80, 16]->[bz*5, 512, 1, 1]
|
|
|
if a_alpha != 1.:
|
|
|
audio_embedding *= a_alpha #放大音频强度
|
|
|
|
|
|
feats = []
|
|
|
x = face_sequences
|
|
|
for f in self.face_encoder_blocks:
|
|
|
x = f(x)
|
|
|
feats.append(x)
|
|
|
|
|
|
x = audio_embedding
|
|
|
for f in self.face_decoder_blocks:
|
|
|
x = f(x)
|
|
|
try:
|
|
|
x = torch.cat((x, feats[-1]), dim=1)
|
|
|
except Exception as e:
|
|
|
print(x.size())
|
|
|
print(feats[-1].size())
|
|
|
raise e
|
|
|
|
|
|
feats.pop()
|
|
|
|
|
|
x = self.output_block(x) #[bz*5, 80, 256, 256]->[bz*5, 3, 256, 256]
|
|
|
|
|
|
if input_dim_size > 4: #[bz*5, 3, 256, 256]->[B, 3, 5, 256, 256]
|
|
|
x = torch.split(x, B, dim=0)
|
|
|
outputs = torch.stack(x, dim=2)
|
|
|
|
|
|
else:
|
|
|
outputs = x
|
|
|
|
|
|
return outputs
|
|
|
|
|
|
|
|
|
class Wav2Lip_disc_qual(nn.Module):
|
|
|
def __init__(self):
|
|
|
super(Wav2Lip_disc_qual, self).__init__()
|
|
|
|
|
|
self.face_encoder_blocks = nn.ModuleList([
|
|
|
nn.Sequential(nonorm_Conv2d(3, 32, kernel_size=7, stride=1, padding=3)),
|
|
|
|
|
|
nn.Sequential(nonorm_Conv2d(32, 64, kernel_size=5, stride=(1, 2), padding=2),
|
|
|
nonorm_Conv2d(64, 64, kernel_size=5, stride=1, padding=2)),
|
|
|
|
|
|
nn.Sequential(nonorm_Conv2d(64, 128, kernel_size=5, stride=2, padding=2),
|
|
|
nonorm_Conv2d(128, 128, kernel_size=5, stride=1, padding=2)),
|
|
|
|
|
|
nn.Sequential(nonorm_Conv2d(128, 256, kernel_size=5, stride=2, padding=2),
|
|
|
nonorm_Conv2d(256, 256, kernel_size=5, stride=1, padding=2)),
|
|
|
|
|
|
nn.Sequential(nonorm_Conv2d(256, 512, kernel_size=3, stride=2, padding=1),
|
|
|
nonorm_Conv2d(512, 512, kernel_size=3, stride=1, padding=1)),
|
|
|
|
|
|
nn.Sequential(nonorm_Conv2d(512, 512, kernel_size=3, stride=2, padding=1),
|
|
|
nonorm_Conv2d(512, 512, kernel_size=3, stride=1, padding=1), ),
|
|
|
|
|
|
nn.Sequential(nonorm_Conv2d(512, 512, kernel_size=3, stride=2, padding=1),
|
|
|
nonorm_Conv2d(512, 512, kernel_size=3, stride=1, padding=1), ),
|
|
|
|
|
|
nn.Sequential(nonorm_Conv2d(512, 512, kernel_size=4, stride=1, padding=0),
|
|
|
nonorm_Conv2d(512, 512, kernel_size=1, stride=1, padding=0)), ])
|
|
|
|
|
|
self.binary_pred = nn.Sequential(nn.Conv2d(512, 1, kernel_size=1, stride=1, padding=0), nn.Sigmoid())
|
|
|
self.label_noise = .0
|
|
|
|
|
|
def get_lower_half(self, face_sequences): #取得输入图片的下半部分。
|
|
|
return face_sequences[:, :, face_sequences.size(2) // 2:]
|
|
|
|
|
|
def to_2d(self, face_sequences): #将输入的图片序列连接起来，形成一个二维的tensor。
|
|
|
B = face_sequences.size(0)
|
|
|
face_sequences = torch.cat([face_sequences[:, :, i] for i in range(face_sequences.size(2))], dim=0)
|
|
|
return face_sequences
|
|
|
|
|
|
def perceptual_forward(self, false_face_sequences): #前传生成图像
|
|
|
false_face_sequences = self.to_2d(false_face_sequences) #[bz, 3, 5, 256, 256]->[bz*5, 3, 256, 256]
|
|
|
false_face_sequences = self.get_lower_half(false_face_sequences)#[bz*5, 3, 256, 256]->[bz*5, 3, 128, 256]
|
|
|
|
|
|
false_feats = false_face_sequences
|
|
|
for f in self.face_encoder_blocks: #[bz*5, 3, 128, 256]->[bz*5, 512, 1, 1]
|
|
|
false_feats = f(false_feats)
|
|
|
|
|
|
return self.binary_pred(false_feats).view(len(false_feats), -1) #[bz*5, 512, 1, 1]->[bz*5, 1, 1]
|
|
|
|
|
|
def forward(self, face_sequences): #前传真值图像
|
|
|
face_sequences = self.to_2d(face_sequences)
|
|
|
face_sequences = self.get_lower_half(face_sequences)
|
|
|
|
|
|
x = face_sequences
|
|
|
for f in self.face_encoder_blocks:
|
|
|
x = f(x)
|
|
|
|
|
|
return self.binary_pred(x).view(len(x), -1)
|