A zero-shot learning framework via cluster-prototype matching

作者：

Highlights：

• A novel Cluster-Prototype Matching (CPM) strategy is proposed to solve the domain shift problem of zero-shot learning.

• The distribution information of samples in the embedding space, i.e., cluster structure, is utilized to assist in zero-shot learning classification.

• The CPM strategy can be applied to most existing ZSL models in a plug-and-play style.

• Extensive experiments show the effectiveness and robustness of the proposed method.

摘要

•A novel Cluster-Prototype Matching (CPM) strategy is proposed to solve the domain shift problem of zero-shot learning.•The distribution information of samples in the embedding space, i.e., cluster structure, is utilized to assist in zero-shot learning classification.•The CPM strategy can be applied to most existing ZSL models in a plug-and-play style.•Extensive experiments show the effectiveness and robustness of the proposed method.

论文关键词：Zero-shot learning,Image classification,Cluster-prototype matching,Domain shift

论文评审过程：Received 31 May 2021, Revised 15 October 2021, Accepted 27 November 2021, Available online 29 November 2021, Version of Record 7 December 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108469