Adaptive fusion of particle filtering and spatio-temporal motion energy for human tracking

Abstract

Object tracking is an active research area nowadays due to its importance in human computer interface, teleconferencing and video surveillance. However, reliable tracking of objects in the presence of occlusions, pose and illumination changes is still a challenging topic. In this paper, we introduce a novel tracking approach that fuses two cues namely colour and spatiotemporal motion energy within a particle filter based framework. We conduct a measure of coherent motion over two image frames, which reveals the spatio-temporal dynamics of the target. At the same time, the importance of both colour and motion energy cues is determined in the stage of reliability evaluation. This determination helps maintain the performance of the tracking system against abrupt appearance changes. Experimental results demonstrate that the proposed method outperforms the other state of the art techniques in the used test datasets.Keywords: Computer vision, object tracking, occlusion, colour, motion energy. * Corresponding author. Tel: +44-28-90971753; fax: +44-28-90971702.Email address: h.zhou@ecit.qub.ac.uk (Huiyu Zhou)Preprint submitted to Pattern Recognition February 2, 2014Figure 1: Flowchart of proposed spatio-temporal based particle filter tracking system.

A powerful approach to real-time mobile objects tracking in crowded environments utilizing 3D video analysis of fixed camera, particle filter and neural network

“…The corners are used as a differentiating feature. A 3D model applied to frames, captured by two cameras may be used to locate walking people [6][7][8][9][10][11][12][13][14][15][16]. Wavelet coefficients could also be applied to track objects in crowded environments.…”

Robust object tracking based upon adaptive multi-cue integration for video surveillance

“…The final likelihood was linear combination of individual likelihood weighted through reliability score. In [33], Huiyu Zhou et al considered color and Spatio-temporal motion energy for object tracking in video sequences. The cues were adaptively combined in particle filter framework.…”

“…We have taken normalized weights for update state after resolving conflicts between multi-cue Eq. (33). In this framework for object representation, the scaling and rotation effect of object are catered to limited extend and tracking results are illustrated in Section 6.…”

A Biologically Inspired Appearance Model for Robust Visual Tracking

“…In these four stages, appearance modeling attracts large attention in the last few decades [7], [17]- [30]. A good appearance model should not only be used to distinguish the target from its background but also be robust against appearance variations due to pose changes, illumination and occlusions.…”