Fine-grained few-shot learning is a difficult task in image classification.The reason is that the discriminative features of fine-grained images are often located in local areas of the image,while most of the existing few-shot learning image classification methods only use top-level features and adopt a single measure.In that way,the local features of the sample cannot be learned well.In response to this problem,ensemble relation network with multi-level measure(ERN-MM)is proposed in this paper.It adds the relation modules in the shallow feature space to compare the similarity between the samples in the local features,and finally integrates the similarity scores from the feature spaces to assign the label of the query samples.So the proposed method ERN-MM can use local details and global information of different grains.Experimental results on different fine-grained datasets show that the proposed method achieves good classification performance and also proves its rationality.

Li Xiaoxu;Qu Xue;Cao Jie

College of Computer and Communication,Lanzhou University of Technology,Lanzhou 730050,China;College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050,China;Engineering Research Center of Urban Railway Transportation of Gansu Province,Lanzhou 730050,China

中国邮电高校学报（英文版）

[1]Li Xiaoxu;Qu Xue;Cao Jie-.Ensemble relation network with multi-level measure)[J].中国邮电高校学报（英文版）,2022(03):15-24,33

ERN

Ensemble,relation,network,multi,level,measure,Fine,grained,few,shot,learning,difficult,task,classification,reason,that,discriminative,features,fine,images,often,located,local,areas,while,most,existing,methods,only,use,top,adopt,single,In,way,cannot,learned,well,response,this,problem,ensemble,MM,proposed,paper,It,adds,modules,shallow,compare,similarity,between,samples,finally,integrates,scores,from,spaces,assign,label,query,So,details,global,information,different,grains,Experimental,results,datasets,show,achieves,good,performance,also,proves,its,rationality

0.54818