With the development of face image syn-thesis and generation technology based on generative ad-versarial networks(GANs),it has become a research hot-spot to determine whether a given face image is natural or generated.However,the generalization capability of the existing algorithms is still to be improved.Therefore,this paper proposes a general algorithm.To do so,firstly,the learning on important local areas,containing many face key-points,is strengthened by combining the global and local features.Secondly,metric learning based on the ArcFace loss is applied to extract common and discrimin-ative features.Finally,the extracted features are fed into the classification module to detect GAN-generated faces.The experiments are conducted on two publicly available natural datasets(CelebA and FFHQ)and seven GAN-generated datasets.Experimental results demonstrate that the proposed algorithm achieves a better generalization performance with an average detection accuracy over 0.99 than the state-of-the-art algorithms.Moreover,the pro-posed algorithm is robust against additional attacks,such as Gaussian blur,and Gaussian noise addition.

Engineering Research Center of Digital Forensics,Ministry of Education,Nanjing University of Information Science and Technology,Nanjing 210044,China;School of Computer,Nanjing University of Information Science and Technology,Nanjing 210044,China;Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology(CICAEET),Nanjing University of Information Science and Technology,Nanjing 210044,China;Warwick Manufacturing Group,University of Warwick,Coventry CV4 7AL,UK;Center for Machine Vision and Signal Analysis,University of Oulu,Oulu 90014,Finland

[1]CHEN Beijing;TAN Weijin;WANG Yiting;ZHAO Guoying-.Distinguishing Between Natural and GAN-Generated Face Images by Combining Global and Local Features)[J].电子学报（英文）,2022(01):59-67

versarial,discrimin

Distinguishing,Between,Natural,Generated,Images,by,Combining,Global,Local,Features,With,development,image,syn,thesis,generation,technology,generative,networks,GANs,has,become,research,hot,spot,determine,whether,given,natural,generated,However,generalization,capability,existing,algorithms,still,improved,Therefore,this,paper,proposes,To,do,so,firstly,learning,important,local,areas,containing,many,key,points,strengthened,combining,global,features,Secondly,metric,ArcFace,loss,applied,common,Finally,extracted,fed,into,classification,module,faces,experiments,conducted,publicly,available,datasets,CelebA,FFHQ,seven,Experimental,results,demonstrate,that,proposed,achieves,better,performance,average,detection,accuracy,than,state,art,Moreover,robust,against,additional,attacks,such,Gaussian,blur,noise

0.62259