Learning deep representation for trajectory clustering. Issue 2 (22nd February 2018)

• Record Type: Journal Article
• Authors: Yao, Di; Zhang, Chao; Zhu, Zhihua; Hu, Qin; Wang, Zheng; Huang, Jianhui; Bi, Jingping
• Other Names: GarcÍa‐RodrÍguez José guestEditor.; Escalera Sergio guestEditor.; Psarrou Alexandra guestEditor.; Guyon Isabelle guestEditor.; Lewis Andrew guestEditor.; Leitner Juxi guestEditor.
• Is Part Of: Expert systems. Volume 35:Issue 2(2018)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2018-02-22
• Subjects: recurrent neural network -- representation learning -- sequence‐to‐sequence learning -- trajectory clustering; Expert systems (Computer science); 006.33
• DOI: 10.1111/exsy.12252
• Languages: English
• ISSNs: 0266-4720

Abstract

Abstract: Trajectory clustering, which aims at discovering groups of similar trajectories, has long been considered as a corner stone task for revealing movement patterns as well as facilitating higher level applications such as location prediction and activity recognition. Although a plethora of trajectory clustering techniques have been proposed, they often rely on spatio‐temporal similarity measures that are not space and time invariant. As a result, they cannot detect trajectory clusters where the within‐cluster similarity occurs in different regions and time periods. In this paper, we revisit the trajectory clustering problem by learning quality low‐dimensional representations of the trajectories. We first use a sliding window to extract a set of moving behaviour features that capture space‐ and time‐invariant characteristics of the trajectories. With the feature extraction module, we transform each trajectory into a feature sequence to describe object movements and further employ a sequence‐to‐sequence auto‐encoder to learn fixed‐length deep representations. The learnt representations robustly encode the movement characteristics of the objects and thus lead to space‐ and time‐invariant clusters. We evaluate the proposed method on both synthetic and real data and observe significant performance improvements over existing methods.