Generating Faces with Deconvolution Networks

用深度学习做人脸合成,website:https://zo7.github.io/blog/2016/09/25/generating-faces.html

受启发于 Learning to Generate Chairs, Tables, and Cars with Convolutional Networks

模型描述

给定一个数据集

包含:c – the one-hot encoding of the model identity

        v – azimuth and elevation of the camera position

       Θ the parameters of additional artificial transformations (增加训练集的数量,减少过拟合)

目标(the RGB output image x, the segmentation mask s)

网络结构

“1s-S-deep” model

生成网络模型由两阶段构成:

1. FC-1 to FC-4 建立一个分享的、高维的隐表达 h(c,v,Θ)

2. FC-5 and uconv-1 to uconv-4 (这部分定义为u)生成outputimage和segmentation mask

这个 deconvolution network类似于 herehere, or here,首先upsample输入,然后convolution。

该模型建立在Keras上。

网络训练

网络参数W

LRGB(squared Euclidean)和Lsegm(squared Euclidean/negative log-likelihood)是损失函数

用更理论的方法生成新模型,训练一个概率生成模型(FC-2)隐状态z:潜在的椅子图像集合

定义 a segmentation mask si under transformation TΘi

定义the pixels in an image xi

log likelihood of an image and its segmentation mask

 网络分析

activating neurons of FC-1 and FC-2 feature maps 见下图(最左边是 setting all neurons of the layer

to zero,其余图像是activating one randomly selected neuron) 并没有太大变化

activating neurons of FC-3 and FC-4 feature maps ,出现视角和类的变化

Images generated from single neurons of the convolutional layers (From top to bottom: uconv-2,

uconv-1, FC-5 of the RGB stream)

接下来,将通过程序进一步理解该模型。

时间: 2024-11-01 10:21:37

Generating Faces with Deconvolution Networks的相关文章

Deformable Convolutional Networks

1 空洞卷积 1.1 理解空洞卷积 在图像分割领域,图像输入到CNN(典型的网络比如FCN)中,FCN先像传统的CNN那样对图像做卷积再pooling,降低图像尺寸的同时增大感受野,但是由于图像分割预测是pixel-wise的输出,所以要将pooling后较小的图像尺寸upsampling到原始的图像尺寸进行预测,之前的pooling操作使得每个pixel预测都能看到较大感受野信息.因此图像分割FCN中有两个关键,一个是pooling减小图像尺寸增大感受野,另一个是upsampling扩大图像尺

CVPR 2017 Paper list

CVPR2017 paper list Machine Learning 1 Spotlight 1-1A Exclusivity-Consistency Regularized Multi-View Subspace Clustering Xiaojie Guo, Xiaobo Wang, Zhen Lei, Changqing Zhang, Stan Z. Li Borrowing Treasures From the Wealthy: Deep Transfer Learning Thro

提高驾驶技术:用GAN去除(爱情)动作片中的马赛克和衣服

作为一名久经片场的老司机,早就想写一些探讨驾驶技术的文章.这篇就介绍利用生成式对抗网络(GAN)的两个基本驾驶技能: 1) 去除(爱情)动作片中的马赛克 2) 给(爱情)动作片中的女孩穿(tuo)衣服 生成式模型 上一篇<用GAN生成二维样本的小例子>中已经简单介绍了GAN,这篇再简要回顾一下生成式模型,算是补全一个来龙去脉. 生成模型就是能够产生指定分布数据的模型,常见的生成式模型一般都会有一个用于产生样本的简单分布.例如一个均匀分布,根据要生成分布的概率密度函数,进行建模,让均匀分布中的样

RNNs学习总结

递归神经网络(recurrent neural networks,RNNs) RNNs的应用主要有两个方面:(1)根据真实世界中发生的相似度对任意的句子进行评分,从而提供一种对语法和语义的正确性进行测量方法,这种模型可以用于机器翻译系统中.(2)预测下一个将要出现的文本,这个应用的价值相对较低. 一.RNNs介绍 在传统的神经网络中,常常假设输入(与输出)之间是互相独立的,然而这在很多应用中是不切实际的,比如要想预测下一个将要出现的文本,那么上下文的作用就不容忽视.RNNs的设计思想就是想要利用

卷积神经网络用于视觉识别Convolutional Neural Networks for Visual Recognition

Table of Contents: Architecture Overview ConvNet Layers Convolutional Layer Pooling Layer Normalization Layer Fully-Connected Layer Converting Fully-Connected Layers to Convolutional Layers ConvNet Architectures Layer Patterns Layer Sizing Patterns C

Classifying plankton with deep neural networks

Classifying plankton with deep neural networks The National Data Science Bowl, a data science competition where the goal was to classify images of plankton, has just ended. I participated with six other members of my research lab, the Reservoir lab o

[C3] Andrew Ng - Neural Networks and Deep Learning

About this Course If you want to break into cutting-edge AI, this course will help you do so. Deep learning engineers are highly sought after, and mastering deep learning will give you numerous new career opportunities. Deep learning is also a new "s

[C6] Andrew Ng - Convolutional Neural Networks

About this Course This course will teach you how to build convolutional neural networks and apply it to image data. Thanks to deep learning, computer vision is working far better than just two years ago, and this is enabling numerous exciting applica

Paper Reading: Perceptual Generative Adversarial Networks for Small Object Detection

Perceptual Generative Adversarial Networks for Small Object Detection 2017-07-11  19:47:46   CVPR 2017 This paper use GAN to handle the issue of small object detection which is a very hard problem in general object detection. As shown in the followin