典型文献
Exploration of the Relation between Input Noise and Generated Image in Generative Adversarial Networks
文献摘要：
In this paper, we propose a hybrid model aiming to map the input noise vector to the label of the generated image by the generative adversarial network (GAN). This model mainly consists of a pre-trained deep convolution generative adversarial network (DCGAN) and a classifier. By using the model, we visualize the distribution of two-dimensional input noise, leading to a specific type of the generated image after each training epoch of GAN. The visualization reveals the distribution feature of the input noise vector and the performance of the generator. With this feature, we try to build a guided generator (GG) with the ability to produce a fake image we need. Two methods are proposed to build GG. One is the most significant noise (MSN) method, and the other utilizes labeled noise. The MSN method can generate images precisely but with less variations. In contrast, the labeled noise method has more variations but is slightly less stable. Finally, we propose a criterion to measure the performance of the generator, which can be used as a loss function to effectively train the network.
文献关键词：
中图分类号：
作者姓名：
Hao-He Liu;Si-Qi Yao;Cheng-Ying Yang;Yu-Lin Wang
作者机构：
School of Computer Science, Northwestern Polytechnical University, Xi'an 710072;International Institute of Service Engineering, Hangzhou Normal University, Hangzhou 311121;Department of Computer Science, University of Taipei, Taipei 10048;Shenzhen Research Institute, Wuhan University, Shenzhen 518057
文献出处：
引用格式：
[1]Hao-He Liu;Si-Qi Yao;Cheng-Ying Yang;Yu-Lin Wang-.Exploration of the Relation between Input Noise and Generated Image in Generative Adversarial Networks)[J].电子科技学刊,2022(01):70-80
A类：
B类：
Exploration,Relation,between,Input,Noise,Generated,Image,Generative,Adversarial,Networks,this,paper,hybrid,model,aiming,map,input,noise,vector,generated,by,generative,adversarial,network,This,mainly,consists,trained,deep,convolution,DCGAN,classifier,By,using,visualize,distribution,dimensional,leading,specific,type,after,each,training,epoch,visualization,reveals,feature,performance,generator,With,try,build,guided,GG,ability,produce,fake,need,Two,methods,are,proposed,One,most,significant,MSN,other,utilizes,labeled,images,precisely,less,variations,contrast,has,more,slightly,stable,Finally,criterion,measure,which,used,loss,function,effectively
AB值：
0.545585
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
