Robust view-invariant multiscale gait recognition
作者：
Highlights：
• The paper proposes a two-phase view-invariant multiscale gait recognition method (VI-MGR).
• VI-MGR is also robust to clothing variation and presence of a carried item.
• Phase 1 determines the matching gallery view of the probe using entropy.
• Phase 2 performs multiscale shape analysis using the Gaussian filter.
• A subject is classified using weighted random subspace learning to avoid overfitting.
摘要
Highlights•The paper proposes a two-phase view-invariant multiscale gait recognition method (VI-MGR).•VI-MGR is also robust to clothing variation and presence of a carried item.•Phase 1 determines the matching gallery view of the probe using entropy.•Phase 2 performs multiscale shape analysis using the Gaussian filter.•A subject is classified using weighted random subspace learning to avoid overfitting.
论文关键词：Gait recognition,Entropy,Gaussian filter,Focus value,Weighted random subspace learning
论文评审过程：Received 10 September 2013, Revised 15 August 2014, Accepted 24 September 2014, Available online 6 October 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.09.022
