Incremental Tensor Subspace Learning and Its Applications to Foreground Segmentation and Tracking

Abstract

Appearance modeling is very important for background modeling and object tracking. Subspace learningbased algorithms have been used to model the appearances of objects or scenes. Current vector subspace-based algorithms cannot effectively represent spatial correlations between pixel values. Current tensor subspace-based algorithms construct an offline representation of image ensembles, and current online tensor subspace learning algorithms cannot be applied to background modeling and object tracking. In this paper, we propose an online tensor subspace learning algorithm which models appearance Electronic supplementary material The online version of this article (changes by incrementally learning a tensor subspace representation through adaptively updating the sample mean and an eigenbasis for each unfolding matrix of the tensor. The proposed incremental tensor subspace learning algorithm is applied to foreground segmentation and object tracking for grayscale and color image sequences. The new background models capture the intrinsic spatiotemporal characteristics of scenes. The new tracking algorithm captures the appearance characteristics of an object during tracking and uses a particle filter to estimate the optimal object state. Experimental evaluations against state-of-the-art algorithms demonstrate the promise and effectiveness of the proposed incremental tensor subspace learning algorithm, and its applications to foreground segmentation and object tracking.

Online learning-based robust visual tracking for autonomous landing of Unmanned Aerial Vehicles

“…G. Li et al [15], T. Wang et al [16], D. Wang et al [17], W. Hu et al [18]) have obtained promising tracking performances. D. Ross et al [19] and F. Yang et al [20] utilized an online incremental learning approach for effectively modelling and updating the tracking target with a low dimensional PCA (i.e.…”

A Note on Background Subtraction by Utilizing a New Tensor Approach

“…The Diffusion Bases (DB) [2] methodology has been adopted by decomposing 3-D data into 2-D plane, which denotes the found out background model. The capability of incremental tensor based background modelling [3] has been investigated with application for foreground segmentation and tracking. Another alternative method versus Principal Component Analysis (PCA) has been utilized by applying the concept of Locality Preserving Projections (LPP) [4], which is called as LoPP.…”

Robust Visual Tracking Using Sparse Discriminative Graph Embedding

“…In fact, the visual tracking problem can be viewed as a two-class image classification problem, that is, it should recognize the tracked target from the similar background samples in tracking procedure. In image recognition, subspace representation can effectively improve both the problem of unclear sparse feature representation and the problem of overfull computation consumption, so it is particularly suitable for a visual tracking environment with realtime requirement [14], [16], [17]. In a two-class classification problem, a linear discriminant analysis (LDA) model had been proved to achieve the discriminative subspace by separating one class from the other.…”