作者:\n\t范宇飞，丁博，何勇军\n

Authors:\n\tFAN Yufei,DING Bo,HE Yongjun\n
摘要:\n\t零样本学习(zero-shot learning, ZSL)致力于在训练期间缺乏不可见类数据的情况下，仍能达到对不可见类别分类的目的。目前在生成式方法中，基于联合生成模型VAEGAN的零样本学习是一个研究热点。在此基础上，提出了一个基于判别器反馈VAEGAN(discriminator feedback VAEGAN，DF-VAEGAN)的零样本图像分类方法。该方法在判别器部分引入了一个反馈模块，在训练阶段可以提升模型整体的性能，在特征生成阶段可以结合生成器共同提升特征生成质量，最终通过高质量的合成特征训练分类器，提高分类准确率。本文还通过解码器重建属性特征，并使用循环一致性损失确保生成特征具备语义一致性。传统ZSL和广义零样本（generalized zero-shot learning, GZSL）图像分类实验展示了本文方法在5个经典数据集上均优于现有方法，在零样本图像分类任务中有效增强了特征合成质量和减少了类别间歧义的目标。\n

Abstract:\n\tZero-shot learning (ZSL) strives to classify unseen categories for which no data is available during training.At present, among generative methods, zero-shot learning based on joint generative model VAEGAN is a research hotspot.On this basis, we propose a zero-shot image classification method based on Discriminator Feedback VAEGAN (DF-VAEGAN).This method introduces a feedback module in the discriminator part, which can improve the overall performance of the model in the training stage.In the feature generation stage, it can be combined with the generator to jointly improve the quality of feature generation.Finally, the classifier is trained through highquality synthetic features to improve classification accuracy.The method also reconstructs attribute features through the decoder and uses a cycle consistency loss to ensure semantic consistency of the generated feature.Experiments on ZSL and generalized zero-shot learning (GZSL) show that our method outperforms existing methods on five classical datasets, effectively enhancing the quality of feature synthesis and reducing the goal of between categories in the zero-shot image classification task.\n


PDF全文下载地址:

可免费Download/下载PDF全文