多层感知器深度卷积生成对抗网络

doi:10.11896/j.issn.1002-137X.2019.09.036

摘要/Abstract

摘要： 生成对抗网络(GAN)是目前图像生成领域中一种新的、有效的训练生成模型方法。深度卷积生成对抗网络(DCGAN)作为GAN的一种延伸,将卷积神经网络引入到生成模型中进行无监督训练。但DCGAN的线性卷积层对于下层数据块是一个广义线性模型,其抽象层次较低,生成的图像质量不高,并且在模型性能度量方面仅以主观的视觉感受来评判图像质量。针对以上问题,文中提出了一种多层感知器深度卷积生成对抗网络(MPDCGAN),采用多层感知器卷积层取代广义线性模型在输入数据上进行卷积,以捕获图像更深层次的特征,并采用定量评估方法Frechet Inception Distance(FID)衡量图像生成质量。在4种基准数据集上的实验结果表明,采用MPDCGAN生成的图像的FID值与图像质量呈负相关关系,且图像生成质量随着FID值的降低得到了进一步的提高。

关键词: FID, 多层感知器, 深度卷积生成对抗网络, 生成对抗网络

Abstract: Generative adversarial network (GAN) is currently a new and effective method for training generative model in image generation.As an extension of GAN,deep convolutional generative adversarial network (DCGAN) introduces convolutional neural networks into the generative model for unsupervised learning.However,the linear convolutional layer of DCGAN is a generalized linear model for the underlying data block.The abstraction level of DCGAN is low and the quality of the generated image is not high.In terms of model performance metrics,image quality is judged only by subjective visual perception.Aiming at the above problems,multi-layer perceptron deep convolutional generative adversarial network (MPDCGAN) was proposed,and the multi-layer perceptron convolutional layer was used to replace the generalized linear model to convolve the input data to capture the deeper features of the image.In order to evaluate the quality of the generated image,a quantitative evaluation method named Frechet Inception Distance (FID) was used.The experimental results on the four benchmark datasets show that the FID value of the image generated by MPDGAN is negatively correlated with the image quality,and the image quality is further improved with the decrease of the FID value.

Key words: Deep convolutional generative adversarial network, Frechet Inception Distance, Generative adversarial network, Multi-layer perceptron

中图分类号:

TP181

王格格, 郭涛, 李贵洋. 多层感知器深度卷积生成对抗网络[J]. 计算机科学, 2019, 46(9): 243-249. https://doi.org/10.11896/j.issn.1002-137X.2019.09.036

WANG Ge-ge, GUO Tao, LI Gui-yang. Multi-layer Perceptron Deep Convolutional Generative Adversarial Network[J]. Computer Science, 2019, 46(9): 243-249. https://doi.org/10.11896/j.issn.1002-137X.2019.09.036

参考文献

[1]LECUN Y,BENDIO Y,HINTON G.Deep Learning[J].Nature,2015,521(7553):436-444.
[2]GOODFELLOW I J,POUGET-ABADIE J,MIRZA M,et al.Generative Adversarial Nets[C]//International Conference on Neural Information Processing Systems.MIT Press,2014:2672-2680.
[3]WANG W L,LI Z R.Advances in Generative Adversarial Network[J].Journal on Communications,2018,39(2):135-148.(in Chinese)王万良,李卓蓉.生成式对抗网络研究进展[J].通信学报,2018,39(2):135-148.
[4]LIN Y L,DAI X Y,LI L,et al.The New Frontier of AI Re-search:Generative Adversarial Networks[J].ACTA Automatica Sinica,2018,44(5):775-792.(in Chinese)林懿伦,戴星原,李力,等.人工智能研究的新前线:生成式对抗网络[J].自动化学报,2018,44(5):775-792.
[5]WANG K F,GOU C,DUAN Y J,et al.Generative Adversarial Networks:the state of the art and beyond[J].ACTA Automatica Sinica,2017,43(3):321-332.(in Chinese)王坤峰,苟超,段艳杰,等.生成式对抗网络GAN的研究进展与展望[J].自动化学报,2017,43(3):321-332.
[6]RATLIFF L J,BURDEN S A,SASTRY S S.Characterizationand Computation of Local Nash Equilibria in Continuous Games[C]//2013 51st Annual Allerton Conference on Communication,Control,and Computing (Allerton).IEEE,2013:917-924.
[7]LEI Y,DING X,WANG S.Visual Tracker using SequentialBayesian Learning:Discriminative,Generative,and Hybrid[J].IEEE Transactions on Systems Man & Cybernetics Part B,2008,38(6):1578-1591.
[8]DINH T B,MEDIONI G.Co-training Framework of Generative and Discriminative Trackers with Partial Occlusion Handling[C]//2011 IEEE Workshop on Applications of Computer Vision (WACV).IEEE,2011:642-649.
[9]ARJOVSKY M,CHINTALA S,BOTTOU L.Wasserstein GAN[J].arXiv:1701.07875,2017.
[10]GULRAJANI I,AHMED F,ARJOVSKY M,et al.ImprovedTraining of Wasserstein Gans [M]//Advances in Neural Information Processing Systems.Berlin:Springer,2017:5767-5777.
[11]MIRZA M,OSINDERO S.Conditional Generative AdversarialNets[J].arXiv:1411.1784,2014.
[12]智能算法研究学习1688.GAN与cGAN [EB/OL].http://www.jinciwei.cn/h303087.html.
[13]CHEN X,DUAN Y,HOUTHOOFT R,et al.Infogan:Inter-pretable Representation Learning by Information Maximizing Generative A-dversarial Nets[M]//Advances in Neural Information Processing Systems.Berlin:Springer,2016:2172-2180.
[14]PERARNAU G.一文帮你发现各种出色的GAN变体[EB/OL].https://chuansongme.com/n/1711678142321.
[15]RADFORD A,METZ L,CHINTALA S.Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks[J].arXiv:1511.06434,2015.
[16]LIN M,CHEN Q,YAN S.Network In Network[J].arXiv:1312.4400,2013.
[17]HEUSEL M,RAMSAUER H,UNTERTHINER T,et al.GANs Trained by A Two Time-scale Update Rule Converge to A Nash equilibrium[J].arXiv:1706.08500,2017.
[18]KRIZHEVAKY A,SUTSKEVER I,HINTON G E.ImageNet Classification with Deep Convolutional Neural Networks[C]//International Conference on Neural Information Processing Systems.Curran Associates Inc.,2012:1097-1105.
[19]BENGIO Y,COURVILLE A,VINCENT P.RepresentationLearning:A Review and New Perspectives[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(8):1798-1828.
[20]知乎.CNN网络的Pooling层有什么用.https://www.zhihu.com/question/36686900.
[21]DENG J,DONG W,SOCHER R,et al.ImageNet:A Large-scale Hierarchical Image Database[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2009.
[22]THEIS L,OORD A,BETHGE M.A Note on the Evaluation of Generative Models[J].arXiv:1511.01844,2015.
[23]SALIMANS T,GOODFELLOW I,ZAREMBA W,et al.Im-proved Techniques for Training Gans[M]//Advances in Neural Information Processing Systems.Berlin:Springs,2016:2234-2242.
[24]BARRATT S,SHARMA R.A Note on the Inception Score[J].arXiv:1801.01973,2018.
[25]百度百科.损失函数[EB/OL].https://baike.baidu.com/item/%E6%8D%9F%E5%A4%B1%E5%87%BD%E6%95%B0/1783236?fr=aladdin.
[26]YUAN M.卷积神经网络的复杂度分析[EB/OL].https://zhuanlan.zhihu.com/p/31575074.

相关文章 15

[1]	张佳, 董守斌. 基于评论方面级用户偏好迁移的跨领域推荐算法 Cross-domain Recommendation Based on Review Aspect-level User Preference Transfer 计算机科学, 2022, 49(9): 41-47. https://doi.org/10.11896/jsjkx.220200131
[2]	孙奇, 吉根林, 张杰. 基于非局部注意力生成对抗网络的视频异常事件检测方法 Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection 计算机科学, 2022, 49(8): 172-177. https://doi.org/10.11896/jsjkx.210600061
[3]	戴朝霞, 李锦欣, 张向东, 徐旭, 梅林, 张亮. 基于DNGAN的磁共振图像超分辨率重建算法 Super-resolution Reconstruction of MRI Based on DNGAN 计算机科学, 2022, 49(7): 113-119. https://doi.org/10.11896/jsjkx.210600105
[4]	尹文兵, 高戈, 曾邦, 王霄, 陈怡. 基于时频域生成对抗网络的语音增强算法 Speech Enhancement Based on Time-Frequency Domain GAN 计算机科学, 2022, 49(6): 187-192. https://doi.org/10.11896/jsjkx.210500114
[5]	徐辉, 康金梦, 张加万. 基于特征感知的数字壁画复原方法 Digital Mural Inpainting Method Based on Feature Perception 计算机科学, 2022, 49(6): 217-223. https://doi.org/10.11896/jsjkx.210500105
[6]	高志宇, 王天荆, 汪悦, 沈航, 白光伟. 基于生成对抗网络的5G网络流量预测方法 Traffic Prediction Method for 5G Network Based on Generative Adversarial Network 计算机科学, 2022, 49(4): 321-328. https://doi.org/10.11896/jsjkx.210300240
[7]	黎思泉, 万永菁, 蒋翠玲. 基于生成对抗网络去影像的多基频估计算法 Multiple Fundamental Frequency Estimation Algorithm Based on Generative Adversarial Networks for Image Removal 计算机科学, 2022, 49(3): 179-184. https://doi.org/10.11896/jsjkx.201200081
[8]	石达, 芦天亮, 杜彦辉, 张建岭, 暴雨轩. 基于改进CycleGAN的人脸性别伪造图像生成模型 Generation Model of Gender-forged Face Image Based on Improved CycleGAN 计算机科学, 2022, 49(2): 31-39. https://doi.org/10.11896/jsjkx.210600012
[9]	唐雨潇, 王斌君. 基于深度生成模型的人脸编辑研究进展 Research Progress of Face Editing Based on Deep Generative Model 计算机科学, 2022, 49(2): 51-61. https://doi.org/10.11896/jsjkx.210400108
[10]	李建, 郭延明, 于天元, 武与伦, 王翔汉, 老松杨. 基于生成对抗网络的多目标类别对抗样本生成算法 Multi-target Category Adversarial Example Generating Algorithm Based on GAN 计算机科学, 2022, 49(2): 83-91. https://doi.org/10.11896/jsjkx.210800130
[11]	谈馨悦, 何小海, 王正勇, 罗晓东, 卿粼波. 基于Transformer交叉注意力的文本生成图像技术 Text-to-Image Generation Technology Based on Transformer Cross Attention 计算机科学, 2022, 49(2): 107-115. https://doi.org/10.11896/jsjkx.210600085
[12]	陈贵强, 何军. 自然场景下遥感图像超分辨率重建算法研究 Study on Super-resolution Reconstruction Algorithm of Remote Sensing Images in Natural Scene 计算机科学, 2022, 49(2): 116-122. https://doi.org/10.11896/jsjkx.210700095
[13]	蒋宗礼, 樊珂, 张津丽. 基于生成对抗网络和元路径的异质网络表示学习 Generative Adversarial Network and Meta-path Based Heterogeneous Network Representation Learning 计算机科学, 2022, 49(1): 133-139. https://doi.org/10.11896/jsjkx.201000179
[14]	张玮琪, 汤轶丰, 李林燕, 胡伏原. 基于场景图的段落生成序列图像方法 Image Stream From Paragraph Method Based on Scene Graph 计算机科学, 2022, 49(1): 233-240. https://doi.org/10.11896/jsjkx.201100207
[15]	林椹尠, 张梦凯, 吴成茂, 郑兴宁. 利用生成对抗网络的人脸图像分步补全法 Face Image Inpainting with Generative Adversarial Network 计算机科学, 2021, 48(9): 174-180. https://doi.org/10.11896/jsjkx.200800014

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed