Learning meta-knowledge for few-shot image emotion recognition

作者：

Highlights：

• Formalize image emotion classification as a few-shot learning problem.

• Explore both classical and generalized few-shot image emotion classification tasks.

• Propose a hierarchical Bayesian model to capture uncertainty and ambiguity.

• Transfer knowledge from label-rich datasets to those with weak supervision.

摘要

•Formalize image emotion classification as a few-shot learning problem.•Explore both classical and generalized few-shot image emotion classification tasks.•Propose a hierarchical Bayesian model to capture uncertainty and ambiguity.•Transfer knowledge from label-rich datasets to those with weak supervision.

论文关键词：Image emotion,Meta-learning,Few-shot learning,Transfer learning,Bayesian learning

论文评审过程：Received 3 June 2020, Revised 23 September 2020, Accepted 6 November 2020, Available online 11 November 2020, Version of Record 24 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114274