Cross modal similarity learning with active queries

作者：

Highlights：

• A novel approach COSLAQ for cross modal similarity learning with active queries is proposed.

• Disagreement-based active query strategy explores the most valuable supervised information.

• Uncertainty of similarity learning model is utilized to avoid querying outliers and noises.

摘要

•A novel approach COSLAQ for cross modal similarity learning with active queries is proposed.•Disagreement-based active query strategy explores the most valuable supervised information.•Uncertainty of similarity learning model is utilized to avoid querying outliers and noises.

论文关键词：Active learning,Cross modal similarity learning,Metric learning

论文评审过程：Received 14 November 2016, Revised 24 March 2017, Accepted 13 May 2017, Available online 18 May 2017, Version of Record 21 November 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.05.011