Designing a symmetric classifier for image annotation using multi-layer sparse coding
作者：
Highlights：
• Proposed a novel annotation system which performs coarse-to-fine labeling.
• System finds thematic relations among images, useful for database organization.
• Performance balanced in terms of precision and recall, and better than old systems.
• High precision for common and rare words, rare words have more information content.
• Many old systems are precise for common words only.
摘要
•Proposed a novel annotation system which performs coarse-to-fine labeling.•System finds thematic relations among images, useful for database organization.•Performance balanced in terms of precision and recall, and better than old systems.•High precision for common and rare words, rare words have more information content.•Many old systems are precise for common words only.
论文关键词：Automatic image annotation,Sparse coding,Symmetric classifier response
论文评审过程：Received 28 June 2016, Revised 14 July 2017, Accepted 10 November 2017, Available online 20 November 2017, Version of Record 25 November 2017.
论文官网地址：https://doi.org/10.1016/j.imavis.2017.11.002
