Decision-level fusion for single-view gait recognition with various carrying and clothing conditions

作者：

Highlights：

• Two new gait representations, AFI and EMAEI, are proposed and evaluated.

• New tensorial gait representation technique is proposed.

• Energy-based gait approaches can perform very well.

• Fusion is recommended in gait recognition, especially decision-level fusion.

• Weighted fusion can generally perform better than unweighted fusion.

摘要

•Two new gait representations, AFI and EMAEI, are proposed and evaluated.•New tensorial gait representation technique is proposed.•Energy-based gait approaches can perform very well.•Fusion is recommended in gait recognition, especially decision-level fusion.•Weighted fusion can generally perform better than unweighted fusion.

论文关键词：Biometrics,Gait recognition,Decision-level fusion,Accumulated prediction image,Accumulated flow image,Edge-masked active energy image,Multilinear subspace learning

论文评审过程：Received 17 March 2015, Revised 29 December 2016, Accepted 8 February 2017, Available online 24 February 2017, Version of Record 14 March 2017.

论文官网地址：https://doi.org/10.1016/j.imavis.2017.02.004