112 lines
4.2 KiB
Python
112 lines
4.2 KiB
Python
import math
|
|
import torch.nn as nn
|
|
import models.basicblock as B
|
|
|
|
|
|
"""
|
|
# --------------------------------------------
|
|
# modified SRResNet
|
|
# -- MSRResNet_prior (for DPSR)
|
|
# --------------------------------------------
|
|
References:
|
|
@inproceedings{zhang2019deep,
|
|
title={Deep Plug-and-Play Super-Resolution for Arbitrary Blur Kernels},
|
|
author={Zhang, Kai and Zuo, Wangmeng and Zhang, Lei},
|
|
booktitle={IEEE Conference on Computer Vision and Pattern Recognition},
|
|
pages={1671--1681},
|
|
year={2019}
|
|
}
|
|
@inproceedings{wang2018esrgan,
|
|
title={Esrgan: Enhanced super-resolution generative adversarial networks},
|
|
author={Wang, Xintao and Yu, Ke and Wu, Shixiang and Gu, Jinjin and Liu, Yihao and Dong, Chao and Qiao, Yu and Change Loy, Chen},
|
|
booktitle={European Conference on Computer Vision (ECCV)},
|
|
pages={0--0},
|
|
year={2018}
|
|
}
|
|
@inproceedings{ledig2017photo,
|
|
title={Photo-realistic single image super-resolution using a generative adversarial network},
|
|
author={Ledig, Christian and Theis, Lucas and Husz{\'a}r, Ferenc and Caballero, Jose and Cunningham, Andrew and Acosta, Alejandro and Aitken, Andrew and Tejani, Alykhan and Totz, Johannes and Wang, Zehan and others},
|
|
booktitle={IEEE conference on computer vision and pattern recognition},
|
|
pages={4681--4690},
|
|
year={2017}
|
|
}
|
|
# --------------------------------------------
|
|
"""
|
|
|
|
|
|
# --------------------------------------------
|
|
# MSRResNet super-resolver prior for DPSR
|
|
# https://github.com/cszn/DPSR
|
|
# https://github.com/cszn/DPSR/blob/master/models/network_srresnet.py
|
|
# --------------------------------------------
|
|
class MSRResNet_prior(nn.Module):
|
|
def __init__(self, in_nc=4, out_nc=3, nc=96, nb=16, upscale=4, act_mode='R', upsample_mode='upconv'):
|
|
super(MSRResNet_prior, self).__init__()
|
|
n_upscale = int(math.log(upscale, 2))
|
|
if upscale == 3:
|
|
n_upscale = 1
|
|
|
|
m_head = B.conv(in_nc, nc, mode='C')
|
|
|
|
m_body = [B.ResBlock(nc, nc, mode='C'+act_mode+'C') for _ in range(nb)]
|
|
m_body.append(B.conv(nc, nc, mode='C'))
|
|
|
|
if upsample_mode == 'upconv':
|
|
upsample_block = B.upsample_upconv
|
|
elif upsample_mode == 'pixelshuffle':
|
|
upsample_block = B.upsample_pixelshuffle
|
|
elif upsample_mode == 'convtranspose':
|
|
upsample_block = B.upsample_convtranspose
|
|
else:
|
|
raise NotImplementedError('upsample mode [{:s}] is not found'.format(upsample_mode))
|
|
if upscale == 3:
|
|
m_uper = upsample_block(nc, nc, mode='3'+act_mode)
|
|
else:
|
|
m_uper = [upsample_block(nc, nc, mode='2'+act_mode) for _ in range(n_upscale)]
|
|
|
|
H_conv0 = B.conv(nc, nc, mode='C'+act_mode)
|
|
H_conv1 = B.conv(nc, out_nc, bias=False, mode='C')
|
|
m_tail = B.sequential(H_conv0, H_conv1)
|
|
|
|
self.model = B.sequential(m_head, B.ShortcutBlock(B.sequential(*m_body)), *m_uper, m_tail)
|
|
|
|
def forward(self, x):
|
|
x = self.model(x)
|
|
return x
|
|
|
|
|
|
|
|
class SRResNet(nn.Module):
|
|
def __init__(self, in_nc=3, out_nc=3, nc=64, nb=16, upscale=4, act_mode='R', upsample_mode='upconv'):
|
|
super(SRResNet, self).__init__()
|
|
n_upscale = int(math.log(upscale, 2))
|
|
if upscale == 3:
|
|
n_upscale = 1
|
|
|
|
m_head = B.conv(in_nc, nc, mode='C')
|
|
|
|
m_body = [B.ResBlock(nc, nc, mode='C'+act_mode+'C') for _ in range(nb)]
|
|
m_body.append(B.conv(nc, nc, mode='C'))
|
|
|
|
if upsample_mode == 'upconv':
|
|
upsample_block = B.upsample_upconv
|
|
elif upsample_mode == 'pixelshuffle':
|
|
upsample_block = B.upsample_pixelshuffle
|
|
elif upsample_mode == 'convtranspose':
|
|
upsample_block = B.upsample_convtranspose
|
|
else:
|
|
raise NotImplementedError('upsample mode [{:s}] is not found'.format(upsample_mode))
|
|
if upscale == 3:
|
|
m_uper = upsample_block(nc, nc, mode='3'+act_mode)
|
|
else:
|
|
m_uper = [upsample_block(nc, nc, mode='2'+act_mode) for _ in range(n_upscale)]
|
|
|
|
H_conv0 = B.conv(nc, nc, mode='C'+act_mode)
|
|
H_conv1 = B.conv(nc, out_nc, bias=False, mode='C')
|
|
m_tail = B.sequential(H_conv0, H_conv1)
|
|
|
|
self.model = B.sequential(m_head, B.ShortcutBlock(B.sequential(*m_body)), *m_uper, m_tail)
|
|
|
|
def forward(self, x):
|
|
x = self.model(x)
|
|
return x |