QFuse: Online learning framework for adaptive biometric system

作者：

Highlights：

• Image and modality-specific quality assessment of multi-modal images.

• Quality based context switching framework for dynamic classifier selection.

• Updating the context switching rules in an online manner with new incremental users.

• Experimental results on the WVU and a real world multimodal databases show improved performance.

摘要

Highlights•Image and modality-specific quality assessment of multi-modal images.•Quality based context switching framework for dynamic classifier selection.•Updating the context switching rules in an online manner with new incremental users.•Experimental results on the WVU and a real world multimodal databases show improved performance.

论文关键词：Biometrics,Classifier selection,Online learning,Context switching

论文评审过程：Received 15 November 2013, Revised 5 March 2015, Accepted 2 May 2015, Available online 13 May 2015, Version of Record 16 July 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.05.002