Learning discriminative and representative feature with cascade GAN for generalized zero-shot learning

作者：

Highlights：

• Generating more representative latent and visual features to alleviate domain shift.

• Features of different categories are more discriminative through orthogonal projection.

• It can outperform the SOTA methods on five popular datasets.

摘要

•Generating more representative latent and visual features to alleviate domain shift.•Features of different categories are more discriminative through orthogonal projection.•It can outperform the SOTA methods on five popular datasets.

论文关键词：Generalized zero-shot learning,Generative models,Orthogonality,Cascade GAN

论文评审过程：Received 4 July 2021, Revised 14 November 2021, Accepted 15 November 2021, Available online 30 November 2021, Version of Record 10 December 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107780