Bias alleviating generative adversarial network for generalized zero-shot classification

作者：

Highlights：

• Train the generator by generating seen and unseen samples, simultaneously.

• Make the generated unseen visual prototypes similar to the nearest real cluster centers.

• Preserve the semantic relationships of the seen and the generated unseen visual prototypes.

• The proposed method achieves state-of-the-art results.

摘要

•Train the generator by generating seen and unseen samples, simultaneously.•Make the generated unseen visual prototypes similar to the nearest real cluster centers.•Preserve the semantic relationships of the seen and the generated unseen visual prototypes.•The proposed method achieves state-of-the-art results.

论文关键词：Generalized zero shot classification,Generative adversarial network,Unseen visual prototypes,Cluster centers,Semantic relationships

论文评审过程：Received 3 May 2020, Revised 20 October 2020, Accepted 22 November 2020, Available online 28 November 2020, Version of Record 2 December 2020.

论文官网地址：https://doi.org/10.1016/j.imavis.2020.104077