Algorithm 1: Adversarial training of refiner net-
work R
θ
Input: Sets of synthetic images x
i
∈ X , and real
images y
j
∈ Y, max number of steps (T ),
number of discriminator network updates
per step (K
d
), number of generative
network updates per step (K
g
).
Output: ConvNet model R
θ
.
for t = 1, . . . , T do
for k = 1, . . . , K
g
do
1. Sample a mini-batch of synthetic images
x
i
.
2. Update θ by taking a SGD step on
mini-batch loss L
R
(θ) in (4) .
end
for k = 1, . . . , K
d
do
1. Sample a mini-batch of synthetic images
x
i
, and real images y
j
.
2. Compute
˜
x
i
= R
θ
(x
i
) with current θ.
3. Update φ by taking a SGD step on
mini-batch loss L
D
(φ) in (2).
end
end
Discriminator
D
Probability mapInput image
w
h
Figure 3. Illustration of local adversarial loss. The discrimina-
tor network outputs a w × h probability map. The adversarial
loss function is the sum of the cross-entropy losses over the
local patches.
loss function (1) used in our implementation is:
L
R
(θ) = −
X
i
log(1 − D
φ
(R
θ
(x
i
)))
+λkR
θ
(x
i
) − x
i
k
1
, (4)
where k.k
1
is `
1
norm. We implement R
θ
as a fully con-
volutional neural net without striding or pooling. This
modifies the synthetic image on a pixel level, rather
than holistically modifying the image content as in e.g.
a fully connected encoder network, and preserves the
global structure and the annotations. We learn the refiner
and discriminator parameters by minimizing L
R
(θ) and
L
D
(φ) alternately. While updating the parameters of
R
θ
, we keep φ fixed, and while updating D
φ
, we fix θ.
We summarize this training procedure in Algorithm 1.
2.2. Local Adversarial Loss
Another key requirement for the refiner network is
that it should learn to model the real image characteris-
tics without introducing any artifacts. When we train a
Buffer of
refined images
Refined images
with current
Refined Real
Mini-batch for
D
R
Figure 4. Illustration of using a history of refined images. See
text for details.
single strong discriminator network, the refiner network
tends to over-emphasize certain image features to fool
the current discriminator network, leading to drifting
and producing artifacts. A key observation is that any
local patch we sample from the refined image, should
have similar statistics to a real image patch. Therefore,
rather than defining a global discriminator network, we
can define discriminator network that classifies all local
image patches separately. This not only limits the re-
ceptive field, and hence the capacity of the discriminator
network, but also provides many samples per image for
learning the discriminator network. This also improves
training of the refiner network because we have multiple
‘realism loss’ values per image.
In our implementation, we design the discriminator
D to be a fully convolutional network that outputs w ×
h dimensional probability map of patches belonging to
fake class, where w × h are the number of local patches
in the image. While training the refiner network, we sum
the cross-entropy loss values over w × h local patches,
as illustrated in Figure 3.
2.3. Updating Discriminator using a History of
Refined Images
Another problem of adversarial training is that the
discriminator network only focuses on the latest refined
images. This may cause (i) diverging of the adversar-
ial training, and (ii) the refiner network re-introducing
the artifacts that the discriminator has forgotten about.
Any refined image generated by the refiner network at
any time during the entire training procedure is a ‘fake’
image for the discriminator. Hence, the discriminator
should be able to classify all these images as fake. Based
on this observation, we introduce a method to improve
the stability of adversarial training by updating the dis-
criminator using a history of refined images, rather than
only the ones in the current mini-batch. We slightly
modify Algorithm 1 to have a buffer of refined images
generated by previous networks. Let B be the size of the
buffer and b be the mini-batch size used in Algorithm 1.
剩余15页未读，继续阅读
r1275805710
- 粉丝: 0
- 资源: 1
我的内容管理
展开
最新资源
- 前端面试必问：真实项目经验大揭秘
- 永磁同步电机二阶自抗扰神经网络控制技术与实践
- 基于HAL库的LoRa通讯与SHT30温湿度测量项目
- avaWeb-mast推荐系统开发实战指南
- 慧鱼SolidWorks零件模型库：设计与创新的强大工具
- MATLAB实现稀疏傅里叶变换(SFFT)代码及测试
- ChatGPT联网模式亮相，体验智能压缩技术.zip
- 掌握进程保护的HOOK API技术
- 基于.Net的日用品网站开发：设计、实现与分析
- MyBatis-Spring 1.3.2版本下载指南
- 开源全能媒体播放器：小戴媒体播放器2 5.1-3
- 华为eNSP参考文档：DHCP与VRP操作指南
- SpringMyBatis实现疫苗接种预约系统
- VHDL实现倒车雷达系统源码免费提供
- 掌握软件测评师考试要点：历年真题解析
- 轻松下载微信视频号内容的新工具介绍
资源上传下载、课程学习等过程中有任何疑问或建议，欢迎提出宝贵意见哦~我们会及时处理！
点击此处反馈
