The ordinal relation preserving binary codes
作者：
Highlights：
• An effective and efficient binary coding framework which enjoys the merits of lookup-based and hamming-based methods is proposed.
• Relative error is proposed to preserve the relative sensitivity.
• The effective tangent planes are adopted as hashing functions to avoid computing the complex classification problem.
• The encoding error guarantees that the selected hashing functions have good performance.
摘要
Author-Highlights•An effective and efficient binary coding framework which enjoys the merits of lookup-based and hamming-based methods is proposed.•Relative error is proposed to preserve the relative sensitivity.•The effective tangent planes are adopted as hashing functions to avoid computing the complex classification problem.•The encoding error guarantees that the selected hashing functions have good performance.
论文关键词：Hashing algorithm,Binary codes,Approximate nearest neighbor search,Image retrieval
论文评审过程：Received 30 September 2014, Revised 26 January 2015, Accepted 16 February 2015, Available online 25 February 2015, Version of Record 17 June 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.02.011
