Feature selection for low bit rate mobile augmented reality applications

作者：

Highlights：

• Novel relevance-based feature selection for low bit rate MAR applications.

• Proposed relevance metrics select significant features for content retrieval.

• Proposed methods achieve superior retrieval performance under low bit rate.

• Proposed methods improve retrieval accuracy for different detectors.

• Feature selection should be applied to the distorted image for high accuracy.

摘要

•Novel relevance-based feature selection for low bit rate MAR applications.•Proposed relevance metrics select significant features for content retrieval.•Proposed methods achieve superior retrieval performance under low bit rate.•Proposed methods improve retrieval accuracy for different detectors.•Feature selection should be applied to the distorted image for high accuracy.

论文关键词：Feature selection,Mobile augmented reality,Retrieval accuracy,Low bit-rate transmission

论文评审过程：Received 20 October 2014, Revised 20 March 2015, Accepted 10 June 2015, Available online 27 June 2015, Version of Record 10 July 2015.

论文官网地址：https://doi.org/10.1016/j.image.2015.06.008