The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
At present, the effective tracking of pedestrians is still a challenging task due to factors such as illumination change, pose variation, motion blur and occlusion. In this paper, we propose a simple and effective tracking algorithm which exploits the spatio-temporal context. Based on a existing Bayesian framework, we take full advantage of the relevance of the region of interest to its local context,...
Video object tracking has been a challenging task in computer vision based applications. Most of state-of-the-art tracking methods rely on convolutional neural network to extract features, and then employ observation model to locate target. Recent studies indicate that convolutional feature maps are noisy and much of the activations are not related to tracking task. Moreover, it will increase computation...
Random forest has emerged as a powerful classification technique with promising results in various vision tasks including image classification, pose estimation and object detection. However, current techniques have shown little improvements in visual tracking as they mostly rely on piece wise orthogonal hyperplanes to create decision nodes and lack a robust incremental learning mechanism that is much...
In this paper, we propose a multi-task correlation particle filter (MCPF) for robust visual tracking. We first present the multi-task correlation filter (MCF) that takes the interdependencies among different features into account to learn correlation filters jointly. The proposed MCPF is designed to exploit and complement the strength of a MCF and a particle filter. Compared with existing tracking...
The pantograph-catenary interaction is an important index of railway safety, which is always evaluated and analyzed by the position of the contact points. The tracking and locating of the contact points in online monitoring has always been an urgent problem due to the high speed of the train operation and various complex backgrounds. In this paper, we proposed an practical online tracking method for...
Most video tracking algorithms including L1 tracker often fail to track correctly under adverse conditions such as object occlusion, disappearance, etc. To address this issue, we propose an improved L1 tracker algorithm called Tracker-2, based on what we call the expanded template which includes the reference template and trail template. The reference template keeps the original features of the target...
We propose an extremely simple but effective regularization technique of convolutional neural networks (CNNs), referred to as BranchOut, for online ensemble tracking. Our algorithm employs a CNN for target representation, which has a common convolutional layers but has multiple branches of fully connected layers. For better regularization, a subset of branches in the CNN are selected randomly for...
In recent years, Discriminative Correlation Filter (DCF) based methods have significantly advanced the state-of-the-art in tracking. However, in the pursuit of ever increasing tracking performance, their characteristic speed and real-time capability have gradually faded. Further, the increasingly complex models, with massive number of trainable parameters, have introduced the risk of severe over-fitting...
For the pedestrian tracking failure problems of the multiple concurrent particle filter with partial occlusion in active safety, this paper proposed a fuzzy decision algorithm to decrease the error probability when the multiple deformable parts are used to tracking an up-right person in the video frames. This algorithm applies the multiple adaptive color-based particle filter trackers to trace the...
Convolutional neural network (CNN) has drawn increasing interest in visual tracking owing to its powerfulness in feature extraction. Most existing CNN-based trackers treat tracking as a classification problem. However, these trackers are sensitive to similar distractors because their CNN models mainly focus on inter-class classification. To address this problem, we use self-structure information of...
Video surveillance systems are now widely deployed to improve our lives by enhancing safety, security, health monitoring and business intelligence. This has motivated extensive research into automated video analysis. Nevertheless, there is a gap between the focus of contemporary research, and the needs of end users of video surveillance systems. Many existing benchmarks and methodologies focus on...
The paper addresses the problem of distributed sensor fusion in the framework of random finite set. The Generalized Covariance Intersection (GCI) rule of multi-target densities is extensively used in multi-target Bayesian filtering scheme. But there are two problems in GCI which are unreasonable design of fusion weight and unable to tackle informative differentiation. In order to get rid of the bad...
Target tracking is an important part of many applications in computer vision. With continuous researching and developing, more and more advanced tracking algorithms have been put forward. Therefore, a benchmark and dataset used to evaluate the most advanced tracking algorithms are needed. One of the most widely used benchmarks is Online Tracking Benchmark (OTB). This benchmark has a clear statement...
To address multi-sensor robust track-to-track association in the presence of sensor biases and missed detections, where sensors biases is time-varying and non-uniform, the target of different sensors is non-identical, the robust track-to-track association algorithm based on t-distribution mixture model is proposed. The robust track-to-track association problem is turned into the non-rigid point matching...
Data association, which could be categorized into offline approaches and the online counterparts, is a crucial part of a multi-object tracker in the tracking-by-detection framework. On the one hand, classical offline data association methods exploit all the video data and have high computation cost, which makes them unscalable to long-term offline video data. On the other hand, online approaches have...
Nowadays, HOG (Histogram of Gradient) feature is extracted from the objects and using it in the classification tasks among the many visual application systems such as object tracking, action recognition and automated video surveillance. Most techniques of extraction HOG feature are based on cells and blocks. Although the HOG feature on cell and block are being robust for current visual systems, the...
In this report, we propose a novel robust tumor tracking method for ultrasound guided RFA (radiofrequency ablation) treatments. Organ deformations seriously deteriorate the tracking performance. To cope with this problem, we propose a novel motion tracking method using dynamic templates. Our method achieves stable tracking by selecting templates automatically based on the texture features of ultrasound...
A video segmentation method based on strong target constrained video saliency (STCVS) is proposed in this paper. In order to detect the salient region fast and effectively, the proposed STCVS is extracted based on the extension of image saliency by enforcing the salient region constrained with the location, scale and color model of the target. Besides, according to the results of STCVS, the super-pixel...
In this paper, we propose a Local Soft Similarity based on Soft Cosine Measure (L3SCM) and then we incorporate it into visual tracking framework. Firstly, we present the soft cosine measure that measures the soft similarity between two vectors of features by taking into consideration similarities of pairs of features. Secondly, we apply this soft similarity in the observation model component of the...
Although the classic TLD (tracking-learning-detection) target tracking algorithm can track a single target robustly in a long period, its real-time performance is poor. CT (Compressing Tracking) real-time tracking algorithm is real-time and efficient, but it cannot accurately track the scale-changing targets. Aiming at handling both methods' shortcomings, this paper proposes an improved TLD tracking...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.