其他分享
首页 > 其他分享> > Attribute-Driven Spontaneous Motion in Unpaired Image Translation

Attribute-Driven Spontaneous Motion in Unpaired Image Translation

作者:互联网

解决的问题

前人的做法:

Ÿ   Success of image translation methods mostly imposes the requirement of working on aligned or similar domains for texture or appearance transform.

Ÿ   The building blocks of these networks, such as convolution/deconvolution layers and activation functions, are spatially corresponding.

问题:

Ÿ   As shown in Fig. 1(a-d), artifacts or ghosting could appear when nonsmile and smile faces are not geometrically aligned in image space.

大白话解释一段我的理解:

       非成对图像翻译,比如非笑脸图(小明,男)要翻译成笑脸图(小明,男)。

       输入网络的图,非笑脸图(小明,男)和笑脸图(大红唇的小红,女),经过一系列的convolution/deconvolution layers and activation functions,得到小明的五官信息和大红唇小红的笑信息(maybe 嘴角上扬图),重组以后得到,大红唇的笑脸小明。

       结果图中的大红唇可以看作artifacts or ghosting。

期望的效果:

 

它的具体做法:

 

SPM

 

RM

12个平行采样的残差块,对生成结果进行精细化。

Mask

Ÿ   3 deconvolutional layers to up-sample f into a 1-channel mask m, the same size as the input.

Ÿ   Sigmoid layer is used as the final activation layer to range the output mask in [0; 1].

Ÿ   a regularization term Lm to enforce sparsity of masks in L1-norm: 

实验效果展示

实验详情

Ÿ   数据库:

选择的属性“Smiling’, ‘Arched eyebrow’, ‘Big Nose’, ‘Pointy nose’”,只训练了正脸。

Ÿ   Pytorch

Ÿ   TITAN Xp

Ÿ   训练策略(two-stage training)

ü  Adam,learning rate 1e-4

ü   LR framework, 128 *128 images, batch size 16,  iterations.

ü   Higher resolutions 256*256 or 512*512, batch size 8 , another iterations.

 

 

标签:layers,小明,红唇,Attribute,Image,Motion,笑脸,activation,size
来源: https://www.cnblogs.com/wjjcjj/p/12336730.html