Novelty detection in human tracking based on spatiotemporal oriented energies
作者：
Highlights：
• Exploiting SOE features for modeling occlusion and detecting novelties in videos.
• A “Bayesian model” to determine various states of the target during the tracking.
• A “Novelty Detection” method to prevent template corruption in occlusion and drift.
• Changing the tracking strategy in ‘Full Occlusion’ to a ‘predictive tracking’ method.
• Thorough evaluation in occlusion scenarios on public surveillance datasets.
摘要
Highlights•Exploiting SOE features for modeling occlusion and detecting novelties in videos.•A “Bayesian model” to determine various states of the target during the tracking.•A “Novelty Detection” method to prevent template corruption in occlusion and drift.•Changing the tracking strategy in ‘Full Occlusion’ to a ‘predictive tracking’ method.•Thorough evaluation in occlusion scenarios on public surveillance datasets.
论文关键词：Occlusion modeling,Novelty detection in tracking,Spatiotemporal oriented energy,Image motion analysis,Video surveillance
论文评审过程：Received 13 December 2013, Revised 14 May 2014, Accepted 4 July 2014, Available online 27 July 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.07.004
