Learning binary code for fast nearest subspace search

作者：

Highlights：

• We design a new objective function which not only considers the hamming distances between similar pairs, but also constrains the hamming distances between dissimilar pairs.

• We further extend our method and give a kernelized version to address the circumstances in which the query subspaces have different dimensions.

• Experiments on more public databases with different tasks are presented. The results show that our method outperforms all the state-of-the-art subspace search methods in both searching accuracy and efficiency.

摘要

•We design a new objective function which not only considers the hamming distances between similar pairs, but also constrains the hamming distances between dissimilar pairs.•We further extend our method and give a kernelized version to address the circumstances in which the query subspaces have different dimensions.•Experiments on more public databases with different tasks are presented. The results show that our method outperforms all the state-of-the-art subspace search methods in both searching accuracy and efficiency.

论文关键词：Nearest subspace search,Learning binary code,Hashing,Matrix classifier

论文评审过程：Received 11 January 2019, Revised 12 July 2019, Accepted 8 September 2019, Available online 12 September 2019, Version of Record 19 September 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.107040