We present an online object tracking algorithm based on feature grouping and two-dimensional principal component analysis (2DPCA). Firstly, we introduce regularization into the 2DPCA reconstruction and develop an iterative algorithm to represent an object by 2DPCA bases. Secondly, the object templates are grouped into a more discriminative image and a less discriminative image by computing the variance of the pixels in multiple frames. Then, the projection matrix is learned according to the more discriminative image and the less discriminative image, and the samples are projected. The object tracking results are obtained using Bayesian maximum a posteriori probability estimation. Finally, we employ a template update strategy which combines incremental subspace learning and the error matrix to reduce tracking drift. Compared with other popular methods, our method reduces the computational complexity and is very robust to abnormal changes. Both qualitative and quantitative evaluations on challenging image sequences demonstrate that the proposed tracking algorithm achieves more favorable performance than several state-of-the-art methods.
Online object tracking is a fundamental problem in many computer vision applications such as surveillance, driver assistance systems, and human-computer interactions [
In recent years, as a popular dimensionality reduction and feature extraction technique, linear subspace learning has been successfully used in robust visual tracking. Supervised discriminative methods for classification and regression have also been exploited to solve visual tracking problems. For example, Avidan developed a tracking algorithm that employs the support vector machine (SVM) classifier within an optic flow framework [
Motivated by the abovementioned discussions, we propose an online object tracking algorithm based on feature grouping and 2DPCA. Firstly, we introduce regularization into the 2DPCA reconstruction and develop an iterative algorithm to represent an object by 2DPCA bases. Secondly, the object templates are grouped into a more discriminative image and a less discriminative image by computing the variance of the pixels in multiple frames. Then, the projection matrix is learned according to the more discriminative image and the less discriminative image, and the samples are projected using the projection matrix. The object tracking results are obtained using Bayesian maximum a posteriori probability (MAP) estimation (Figures
Tracking results of test video “Lemming.”
Tracking results of test video “Car4.”
Tracking results of test video “Singer1.”
Principal component analysis (PCA) is a well-established linear dimension-reduction technique, which has been widely used in many areas (such as face recognition [
Given a series of image matrices
Then the coefficient
Let the objective function be
In object tracking problem,
We would like to decompose each template
Denote by
We want to put those pixels having smaller
After decomposing, each template can be written as
Similarly, the average energy of
To preserve the
An equivalent form of (
Apparently, the desired
We can regard object tracking as a hidden state variables’ Bayesian MAP estimation problem in the hidden Markov model; that is, with a set of observed samples
According to the Bayesian theory,
We choose object’s motion affine transformation parameters as state variable
We assume that the state transition model follows the Gaussian distribution; that is,
We use object’s reconstruction error to build observation likelihood model; that is,
In order to show the robustness of the object tracking algorithm based on projection discussed in this paper, we choose several sets of public test videos taken under different environments to test the performance of our algorithm (Figures
The description of test videos.
Name of test videos | Number of frames | Video description |
---|---|---|
Lemming | 1336 | Out-of-plane rotation, scale change, occlusion, and background cluster |
| ||
Car4 | 659 | Illumination variation and scale |
| ||
Singer1 | 351 | Illumination variation and scale change |
Quantitative evaluation in terms of center location error for test video “Lemming.”
Quantitative evaluation in terms of center location error for test video “Car4.”
Quantitative evaluation in terms of center location error for test video “Singer1.”
The implementation of the algorithm is based on Windows operating system. The configuration of the computer is AMD Athlon (TM) X2 Dual Core QL-62 2.00 GHz, 1.74 GB memory. In order to evaluate the performance of the algorithm, we choose six currently most representative and classic tracking algorithms to do the comparison. The six classic algorithms are L1 Tracker [
The average processing speeds of our method for different test videos are listed in Table
The average processing speed of our method.
Name of test videos | The average processing speed (Frame/S) |
---|---|
Lemming | 32.57 |
Car4 | 35.13 |
Singer1 | 29.98 |
This paper presents a robust tracking algorithm via feature grouping and 2DPCA. In this work, we represent the tracked object by using 2DPCA bases and a feature grouping. With the proposed model, we can reduce the effect of abnormal pixels on tracking algorithms. We obtained the tracking result using Bayesian maximum a posteriori probability estimation framework and designed a stable and robust tracker. Then, we explicitly take partial occlusion and misalignment into account for appearance model update and object tracking. Experiments on challenging video clips show that our tracking algorithm performs better than several state-of-the-art algorithms. Our future work will be the generalization of our representation model into other related fields.
This research described in this paper was supported by the Fundamental Research Funds for the Central Universities (DC110321, DC120101132, and DC120101131). This work was supported by project of Liaoning Provincial Department of Education (L2012476 and L2010094) and by National Natural Science Foundation of China (61172058).