生成对抗网络论文

原文地址:https://blog.csdn.net/chognzhihong_seu/article/details/70941000

GAN? — ?Generative Adversarial Networks

3D-GAN? — ?Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling

AC-GAN? —? Conditional Image Synthesis With Auxiliary Classifier GANs

AdaGAN — ?AdaGAN: Boosting Generative Models

AffGAN ?— Amortised MAP Inference for Image Super-resolution

AL-CGAN —? Learning to Generate Images of Outdoor Scenes from Attributes and Semantic Layouts

ALI ?— Adversarially Learned Inference

AMGAN ?— Generative Adversarial Nets with Labeled Data by Activation Maximization

AnoGAN ?—? Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery

ArtGAN? —? ArtGAN: Artwork Synthesis with Conditional Categorial GANs

b-GAN? —? b-GAN: Unified Framework of Generative Adversarial Networks

Bayesian GAN ?—? Deep and Hierarchical Implicit Models

BEGAN ?—? BEGAN: Boundary Equilibrium Generative Adversarial Networks

BiGAN ?—?Adversarial Feature Learning

BS-GAN —? Boundary-Seeking Generative Adversarial Networks

CGAN ?—? Conditional Generative Adversarial Nets

CCGAN?—? Semi-Supervised Learning with Context-Conditional Generative Adversarial Networks

CatGAN ?—? Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks

CoGAN ?—? Coupled Generative Adversarial Networks

Context-RNN-GAN —? Contextual RNN-GANs for Abstract Reasoning Diagram Generation

C-RNN-GAN ?—? C-RNN-GAN: Continuous recurrent neural networks with adversarial training

CVAE-GAN? —? CVAE-GAN: Fine-Grained Image Generation through Asymmetric Training

CycleGAN ?—? Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

DTN ?—? Unsupervised Cross-Domain Image Generation

DCGAN ?—? Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

DiscoGAN ?—? Learning to Discover Cross-Domain Relations with Generative Adversarial Networks

DR-GAN? —? Disentangled Representation Learning GAN for Pose-Invariant Face Recognition

DualGAN ?— ?DualGAN: Unsupervised Dual Learning for Image-to-Image Translation

EBGAN ?—? Energy-based Generative Adversarial Network

f-GAN —? f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization

GAWWN ?—? Learning What and Where to Draw

GoGAN ?—? Gang of GANs: Generative Adversarial Networks with Maximum Margin Ranking

GP-GAN? —? GP-GAN: Towards Realistic High-Resolution Image Blending

IAN ?— Neural Photo Editing with Introspective Adversarial Networks

iGAN ?—? Generative Visual Manipulation on the Natural Image Manifold

IcGAN ?—? Invertible Conditional GANs for image editing

ID-CGAN ?— Image De-raining Using a Conditional Generative Adversarial Network

Improved GAN ?—? Improved Techniques for Training GANs

InfoGAN ?—? InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets

LAPGAN ?—? Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks

LR-GAN ?—? LR-GAN: Layered Recursive Generative Adversarial Networks for Image Generation

LSGAN ?—? Least Squares Generative Adversarial Networks

LS-GAN ?—? Loss-Sensitive Generative Adversarial Networks on Lipschitz Densities

MGAN ?—? Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks

MAGAN ?—? MAGAN: Margin Adaptation for Generative Adversarial Networks

MAD-GAN ?—? Multi-Agent Diverse Generative Adversarial Networks

MalGAN ?—? Generating Adversarial Malware Examples for Black-Box Attacks Based on GAN

MARTA-GAN ?—? Deep Unsupervised Representation Learning for Remote Sensing Images

McGAN ?— McGan: Mean and Covariance Feature Matching GAN

MedGAN ?—? Generating Multi-label Discrete Electronic Health Records using Generative Adversarial Networks

MIX+GAN? —? Generalization and Equilibrium in Generative Adversarial Nets (GANs)

MPM-GAN ?—? Message Passing Multi-Agent GANs

MV-BiGAN ?—? Multi-view Generative Adversarial Networks

pix2pix ?—? Image-to-Image Translation with Conditional Adversarial Networks

PPGN ?—? Plug & Play Generative Networks: Conditional Iterative Generation of Images in Latent Space

PrGAN ?—? 3D Shape Induction from 2D Views of Multiple Objects

RenderGAN ?—? RenderGAN: Generating Realistic Labeled Data

RTT-GAN ?—? Recurrent Topic-Transition GAN for Visual Paragraph Generation

SGAN ?—? Stacked Generative Adversarial Networks

SGAN ?—? Texture Synthesis with Spatial Generative Adversarial Networks

SAD-GAN ?—? SAD-GAN: Synthetic Autonomous Driving using Generative Adversarial Networks

SalGAN ?—? SalGAN: Visual Saliency Prediction with Generative Adversarial Networks

SEGAN ?—? SEGAN: Speech Enhancement Generative Adversarial Network

SeGAN ?—? SeGAN: Segmenting and Generating the Invisible

SeqGAN ?—? SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient

SketchGAN ?—? Adversarial Training For Sketch Retrieval

SL-GAN? —? Semi-Latent GAN: Learning to generate and modify facial images from attributes

Softmax-GAN ?—? Softmax GAN

SRGAN ?—? Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

S²GAN ?—? Generative Image Modeling using Style and Structure Adversarial Networks

SSL-GAN ?—? Semi-Supervised Learning with Context-Conditional Generative Adversarial Networks

StackGAN ?—? StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks

TGAN ?—? Temporal Generative Adversarial Nets

TAC-GAN? —? TAC-GAN?—?Text Conditioned Auxiliary Classifier Generative Adversarial Network

TP-GAN? —? Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis

Triple-GAN —? Triple Generative Adversarial Nets

Unrolled GAN ?—? Unrolled Generative Adversarial Networks

VGAN ?—? Generating Videos with Scene Dynamics

VGAN ?—? Generative Adversarial Networks as Variational Training of Energy Based Models

VAE-GAN —? Autoencoding beyond pixels using a learned similarity metric

VariGAN ?—? Multi-View Image Generation from a Single-View

ViGAN ?—? Image Generation and Editing with Variational Info Generative AdversarialNetworks

WGAN ?—? Wasserstein GAN

WGAN-GP ?—? Improved Training of Wasserstein GANs

WaterGAN ?—? WaterGAN: Unsupervised Generative Network to Enable Real-time Color Correction of Monocular Underwater Images

原文地址:https://www.cnblogs.com/lzhu/p/10480878.html

时间: 2024-11-07 23:08:29

生成对抗网络论文的相关文章

生成对抗网络 Generative Adversarial Networks

转自:https://zhuanlan.zhihu.com/p/26499443 生成对抗网络GAN是由蒙特利尔大学Ian Goodfellow教授和他的学生在2014年提出的机器学习架构. 要全面理解生成对抗网络,首先要理解的概念是监督式学习和非监督式学习.监督式学习是指基于大量带有标签的训练集与测试集的机器学习过程,比如监督式图片分类器需要一系列图片和对应的标签("猫","狗"-),而非监督式学习则不需要这么多额外的工作,它们可以自己从错误中进行学习,并降低未来

《生成对抗网络GAN的原理与应用专题》笔记

视频教程的链接:http://campus.swarma.org/gpac=8 一.什么是GAN 框架简述 GAN全称是Generative Adversarial Nets,中文叫做"生成对抗网络". 在GAN中有2个网络,一个网络用于生成数据,叫做"生成器".另一个网络用于判别生成数据是否接近于真实,叫做"判别器". 下图展示了最简单的GAN的结构以及工作原理. 模型中最左侧的随机向量是为了让生成器产生不同的输出,而增加的扰动项.这些扰动决定

正在涌现的新型神经网络模型:优于生成对抗网络

http://www.17bianji.com/lsqh/35130.html 是以,它会让人想起残差前馈收集(residual feed-forward network),但在实际中,强迫这些收集向前传播误差并不克不及让它们在更高财揭捉习到有效的层次表征.是以,它们不克不及基于更上层的表征来竽暌剐效地履行其它义务,例如分类.瓜分.动作辨认.要明白这些限制,还须要更多的实验. 新一代深度神经收集正在出现.它们演变自前馈模型,之前我们曾作过具体分析,参阅机械之心文┞仿 <重磅 | 神经收集架构演

知物由学 | AI网络安全实战:生成对抗网络

"知物由学"是网易云易盾打造的一个品牌栏目,词语出自汉·王充<论衡·实知>.人,能力有高下之分,学习才知道事物的道理,而后才有智慧,不去求问就不会知道."知物由学"希望通过一篇篇技术干货.趋势解读.人物思考和沉淀给你带来收获的同时,也希望打开你的眼界,成就不一样的你. 以下是正文: 作者:Brad Harris,安全研究员,Brad曾在公共和私营部门的网络和计算机安全领域工作过.他已经完成了从渗透测试到逆向工程到应用研究的所有工作,目前他是IBMX-Fo

生成对抗网络浅析(GAN)

生成对抗网络 ? 顾名思义,生成对抗网络由两个部分构成, 生成器(Generator)和判别器(Discriminator), 两个部件相互博弈,最终达到平衡状态. 基本原理 下面以生成图片为例. G: 生成器 接受一个随机的噪声 z,通过噪声产生目标G(z) D:判别器 判别目标是否是"真实的".输入参数是 x,输出为D(x), 表示是否为真实的概率. ? 训练的过程中, G的目的就是尽量生成真实的图片欺骗 D.而 D的目标就是尽量将 G 生成的图片和真实的图片分离开.这样就是一个博

生成对抗网络GAN

详解一:GAN完整理论推导和实现 详解二:详解生成对抗网络(GAN)原理 原文地址:https://www.cnblogs.com/yunkaiL/p/10952881.html

使用生成对抗网络(GAN)生成手写字

先放结果 这是通过GAN迭代训练30W次,耗时3小时生成的手写字图片效果,大部分的还是能看出来是数字的. 实现原理 简单说下原理,生成对抗网络需要训练两个任务,一个叫生成器,一个叫判别器,如字面意思,一个负责生成图片,一个负责判别图片,生成器不断生成新的图片,然后判别器去判断哪儿哪儿不行,生成器再不断去改进,不断的像真实的图片靠近. 这就如同一个造假团伙一样,A负责生产,B负责就鉴定,刚开始的时候,两个人都是菜鸟,A随便画了一幅画拿给B看,B说你这不行,然后A再改进,当然需要改进的不止A,随着A

PyTorch实现简单的生成对抗网络GAN

生成对抗网络(Generative Adversarial Network, GAN)包括生成网络和对抗网络两部分.生成网络像自动编码器的解码器,能够生成数据,比如生成一张图片.对抗网络用来判断数据的真假,比如是真图片还是假图片,真图片是拍摄得到的,假图片是生成网络生成的. 生成对抗网络就是让生成网络和对抗网络相互竞争,通过生成网络来生成假的数据,对抗网络判别该数据是真是假,最后希望生成网络生成的数据以假乱真骗过判别器. 以下程序主要来自廖星宇的<深度学习之PyTorch>的第六章,本文对原代

利用tensorflow训练简单的生成对抗网络GAN

对抗网络是14年Goodfellow Ian在论文Generative Adversarial Nets中提出来的. 原理方面,对抗网络可以简单归纳为一个生成器(generator)和一个判断器(discriminator)之间博弈的过程.整个网络训练的过程中, 两个模块的分工 判断器,直观来看就是一个简单的神经网络结构,输入就是一副图像,输出就是一个概率值,用于判断真假使用(概率值大于0.5那就是真,小于0.5那就是假) 生成器,同样也可以看成是一个神经网络模型,输入是一组随机数Z,输出是一个