基于线性注意力机制的单样本生成对抗网络研究.

At present, using single-sample training to generate adversarial networks has become the focus of researchers. However, the problems that the model is not easy to converge, the generated image structure collapses, and the training speed is slow still need to be solved urgently. Researchers propose to use a self-attention model in the generative adversarial network to obtain a larger range of samples and improve the quality of the generated images. It is found that using the traditional convolutional self-atention modelcausesawasteofcomputingresourcesduetotheredundancyofinformationinthe attention map. A novel linear attention model is proposed, in which a double normalization method is usedtoaleviatetheproblem of the atent on modelbeingsenstivetoinputfeatures, andanewsingle-samplegenerativeadversarialnetworkmodelisbuiltusingthismodel. In addition, themodelusesresidualnetworkandspectralnormalization methods for stable training, reducing the risk of col apse. A largenumberofexperimentsshowthat, comparedwiththeexistingtraining model, this model has the characteristicsoffasttrainingspeed, highresolutionofgeneratedimages, andobviousimprovementof evaluationindicators. [ABSTRACT FROM AUTHOR]