Robust individual and holistic features for crowd scene classification

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

12 Scopus Citations
View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Original languageEnglish
Pages (from-to)110-120
Journal / PublicationPattern Recognition
Volume58
Online published16 Apr 2016
Publication statusPublished - Oct 2016

Abstract

In this paper, we present an approach that utilizes multiple exemplar agent-based motion models (AMMs) to extract motion features (representing crowd behaviors) from the captured crowd trajectories. In the exemplar-based framework, we propose an iterative optimization algorithm to measure the correlation between any exemplar AMM and the trajectory data. It is based on the Extended Kalman Smoother and KL-divergence. In addition, based on the proposed correlation measure, we introduce the novel individual feature, in combination with the holistic feature, to describe crowd motions. Our results show that the proposed features perform well in classifying real-world crowd scenes.

Research Area(s)

  • Crowd analysis, Crowd scene classification, Crowd modeling, SPARSE REPRESENTATION

Citation Format(s)

Robust individual and holistic features for crowd scene classification. / Liu, Wenxi; Lau, Rynson W. H.; Manocha, Dinesh.
In: Pattern Recognition, Vol. 58, 10.2016, p. 110-120.

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review