scholarly journals Robust Online Object Tracking Based on Feature Grouping and 2DPCA

2013 ◽  
Vol 2013 ◽  
pp. 1-7
Author(s):  
Ming-Xin Jiang ◽  
Jun-Xing Zhang ◽  
Min Li
Keyword(s):  
Object Tracking ◽  
Principal Component ◽  
Subspace Learning ◽  
Tracking Algorithm ◽  
Posteriori Probability ◽  
Error Matrix ◽  
A Posteriori Probability ◽  
Template Update ◽  
Feature Grouping ◽  
Update Strategy

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.

scholarly journals Visual Object Tracking Based on 2DPCA and ML

2013 ◽  
Vol 2013 ◽  
pp. 1-7 ◽  
Author(s):  
Ming-Xin Jiang ◽  
Min Li ◽  
Hong-Yu Wang
Keyword(s):  
Object Tracking ◽  
Principal Component ◽  
Likelihood Estimation ◽  
Subspace Learning ◽  
Tracking Algorithm ◽  
Visual Object ◽  
Visual Object Tracking ◽  
Error Matrix ◽  
Template Update ◽  
Update Strategy

We present a novel visual object tracking algorithm based on two-dimensional principal component analysis (2DPCA) and maximum likelihood estimation (MLE). Firstly, we introduce regularization into the 2DPCA reconstruction and develop an iterative algorithm to represent an object by 2DPCA bases. Secondly, the model of sparsity constrained MLE is established. Abnormal pixels in the samples will be assigned with low weights to reduce their effects on the tracking algorithm. The object tracking results are obtained by using Bayesian maximum a posteriori (MAP) probability estimation. Finally, to further reduce tracking drift, we employ a template update strategy which combines incremental subspace learning and the error matrix. This strategy adapts the template to the appearance change of the target and reduces the influence of the occluded target template as well. 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.

Robust Spatial-Color Feature with New Similarity Measure and Adaptive Template Update for Mean-Shift Tracking

2013 ◽  
Vol 321-324 ◽  
pp. 1021-1029
Author(s):  
Lu Rong Shen ◽  
Xia Bin Dong ◽  
Rui Tao Lu ◽  
Yong Bin Zheng ◽  
Xin Sheng Huang
Keyword(s):  
Object Tracking ◽  
Similarity Measure ◽  
Mean Shift ◽  
Tracking Algorithm ◽  
Experimental Results ◽  
Tracking Accuracy ◽  
Color Feature ◽  
Mean Shift Algorithm ◽  
Template Update ◽  
Update Strategy

In this paper, we analyze the object tracking task of mean-shift algorithm. A spatial-color and similarity based mean-shift tracking algorithm is proposed. The spatial-color feature is used to replace the color histogram, and an enhanced algorithm is derived by adopting a new similarity measure. We also introduce Lucas-Kanade algorithm to design a template update strategy, propose a template update algorithm for mean-shift. Experimental results show that these two improved mean-shift tracking algorithms have high tracking accuracy and good robustness to the change of appearance of the object.

scholarly journals A Novel Incremental Multi-Template Update Strategy for Robust Object Tracking

IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 162668-162682
Author(s):  
Qingsong Xie ◽  
Kewei Liu ◽  
An Zhiyong ◽  
Lei Wang ◽  
Ye Li ◽  
...  
Keyword(s):  
Object Tracking ◽  
Template Update ◽  
Update Strategy

scholarly journals Single object tracking algorithm in occlusion scene based on improved ECO

2021 ◽  
Vol 2132 (1) ◽  
pp. 012010
Author(s):  
Guorong Xie ◽  
Rongqi Jiang ◽  
Yi Qu
Keyword(s):  
Object Tracking ◽  
Success Rate ◽  
Tracking Algorithm ◽  
Single Object ◽  
Feature Map ◽  
Structure Information ◽  
Deep Feature ◽  
Training Samples ◽  
Update Strategy

Abstract To alleviate the occlusion problem in a single object tracking scene, this paper proposes an ECO-MHDU object tracking algorithm with a more powerful anti-occlusion performance based on the ECO tracker. The algorithm first uses the pre-trained MobileNetV3 lightweight backbone network on the ImageNet dataset to replace the ResNet network in the ECO to increase the speed of the algorithm to obtain the shallow and deep feature information of the image, while effectively using the attention mechanism in the MobileNetV3 network to strengthen the algorithm’s ability to extract target features; secondly, use the DropBlock operation on the acquired feature map to generate a random continuous mask on the feature map channel to improve the algorithm’s learning of the global robust spatial structure information; finally, a confidence update strategy is introduced into the GMM sample generation space. To improve the quality of training samples, unreliable tracking states such as confidence detection and occlusion are designed to avoid updating the sample space with damaging information. Compared with the ECO algorithm, the ECO-MHDU algorithm proposed in this paper has a success rate of 68.0% on the occlusion attributes of the OTB100 dataset, which is 2.3% higher than the ECO algorithm, and the ECO-MHDU algorithm also showed the best performance on the entire dataset sequence, with a success rate of 69.3%.

scholarly journals Metode Bayesian Networks untuk Menyelesaikan Occlusion pada Object Tracking

2011 ◽  
Vol 1 (2) ◽  
pp. 89
Author(s):  
Dina Nurul Fitria ◽  
Ikhwan B. Zarkasi ◽  
Rose Maulidiyatul H
Keyword(s):  
Object Recognition ◽  
Bayesian Networks ◽  
Object Tracking ◽  
Video Surveillance ◽  
Posteriori Probability ◽  
Maximum A Posteriori ◽  
A Posteriori ◽  
A Posteriori Probability ◽  
Fixed Area ◽  
Tracking Object

<p style="text-align: justify;" align="center">Banyak cara untuk dapat mendeteksi keamanan sebuah wilayah tertentu. Salah satu cara pengamanan yang bisa digunakan adalah dengan menggunakan pemantauan berbasis video pengawasan (<em>video surveillance</em>). Sebenarnya, video pengawasan sudah banyak digunakan di Indonesia. Tetapi, umumnya video pengawasan ini hanya mampu merekam gambar, tanpa ada kemampuan pintar yakni, <em>object tracking, object recognition</em> dan <em>object analyzing</em>. Sehingga, hasil yang diharapkan kurang maksimal dan belum bisa membantu tugas pengawasan secara keseluruhan. Paper ini bertujuan untuk membuat algoritma dari <em>object tracking</em> yang ada pada video pengawasan sebagai rujukan pengembangan video pengawasan dengan kemampuan <em>object recognition</em> dan <em>object analyzing</em>. Masalah utama yang sering muncul dalam pembuatan <em>object tracking</em> adalah ketika terjadi<em> occlusion</em> (tumpang tindih) antara dua <em>object </em>dalam sebuah frame. Pada saat <em>occlusion</em>, <em>object </em>yang sama pada frame yang berbeda kemungkinan dapat dikenali sebagai<em> object</em> yang berbeda. Sehingga, proses <em>object tracking</em> akan menjadi terganggu. <em>Bayesian Networks</em> memungkinkan untuk membandingkan data yang didapat dari masing-masing <em>object </em>yang ada <em>(likelihood)</em> dengan data awal yang telah dimiliki <em>(prior)</em>, dengan menghitung <em>Maximum A-Posteriori Probability</em>(MAP) yang dimiliki, sehingga <em>object </em>yang sama pada frame yang berbeda tetap akan dikenali sebagai <em>object</em> yang sama</p><h6 style="text-align: center;"><strong> </strong><strong>Abstract</strong></h6><p style="text-align: justify;" align="center">There are many ways/technique to detect the security/safety of fixed area. One of security technique that can be used is by using monitoring based on Video surveillance. In fact, this monitoring video has already been used in Indonesia. But, video surveillance, commonly, just can record images without any smart abilities, such as object tracking, object recognition and object analyzing. So, the expected result is not optimal and still not be able to help monitoring role totally. This research is aimed to make the algorithm of object trackingin video surveillance, in order to be reference for development of video surveillance with ability of object recognition and object analyzing. The main problem that frequently comes up on the making of object tracking is occlusion between two objects in a single frame. When occlusion is happened, same object in different frame probably can be recognized as two different objects. So, the process of object tracking can be disturbed. Bayesian Network is enable to compare data that got from every object (likelihood) with prior data that has already been provided by counting its Maximum A-Posteriori Probability (MAP), so same object in different frame are still be able to be recognized as same object.</p>

Siamese Network Object Tracking Algorithm Combining Attention Mechanism and Correlation Filter Theory

2021 ◽  
Author(s):  
Xiuhua Hu ◽  
Huan Liu ◽  
Yuan Chen ◽  
Yan Hui ◽  
Yingyu Liang ◽  
...  
Keyword(s):  
Object Tracking ◽  
Attention Mechanism ◽  
Tracking Algorithm ◽  
Correlation Filter ◽  
Filter Model ◽  
Filter Theory ◽  
Search Area ◽  
Template Update ◽  
Object Appearance ◽  
Network Object

Aiming to solve the problem of tracking drift during movement, which was caused by the lack of discriminability of the feature information and the failure of a fixed template to adapt to the change of object appearance, the paper proposes an object tracking algorithm combining attention mechanism and correlation filter theory based on the framework of full convolutional Siamese neural networks. Firstly, the apparent information is processed by using the attention mechanism thought, where the object and search area features are optimized according to the spatial attention and channel attention module. At the same time, the cross-attention module is introduced to process the template branch and search area branch, respectively, which makes full use of the diversified context information of the search area. Then, the background perception correlation filter model with scale adaptation and learning rate adjustment is adopted into the model construction, using as a layer in the network model to realize the object template update. Finally, the optimal object location is determined according to the confidence map with similarity calculation. Experimental results show that the designed method in the paper can promote the object tracking performance under various challenging environments effectively; the success rate increases by 16.2%, and the accuracy rate increases by 16%.

High speed long-term visual object tracking algorithm for real robot systems

2021 ◽  
Vol 434 ◽  
pp. 268-284
Author(s):  
Muxi Jiang ◽  
Rui Li ◽  
Qisheng Liu ◽  
Yingjing Shi ◽  
Esteban Tlelo-Cuautle
Keyword(s):  
Object Tracking ◽  
High Speed ◽  
Tracking Algorithm ◽  
Visual Object ◽  
Visual Object Tracking ◽  
Robot Systems ◽  
Real Robot

scholarly journals An Anchor-Free Siamese Network with Multi-Template Update for Object Tracking

Electronics ◽  
2021 ◽  
Vol 10 (9) ◽  
pp. 1067
Author(s):  
Tongtong Yuan ◽  
Wenzhu Yang ◽  
Qian Li ◽  
Yuxia Wang
Keyword(s):  
Object Tracking ◽  
Correlation Energy ◽  
Feature Maps ◽  
Siamese Network ◽  
Template Update ◽  
Free Network ◽  
Multiple Prediction ◽  
Bounding Boxes ◽  
High Level ◽  
Speed And Accuracy

Siamese trackers are widely used in various fields for their advantages of balancing speed and accuracy. Compared with the anchor-based method, the anchor-free-based approach can reach faster speeds without any drop in precision. Inspired by the Siamese network and anchor-free idea, an anchor-free Siamese network (AFSN) with multi-template updates for object tracking is proposed. To improve tracking performance, a dual-fusion method is adopted in which the multi-layer features and multiple prediction results are combined respectively. The low-level feature maps are concatenated with the high-level feature maps to make full use of both spatial and semantic information. To make the results as stable as possible, the final results are obtained by combining multiple prediction results. Aiming at the template update, a high-confidence multi-template update mechanism is used. The average peak to correlation energy is used to determine whether the template should be updated. We use the anchor-free network to implement object tracking in a per-pixel manner, which computes the object category and bounding boxes directly. Experimental results indicate that the average overlap and success rate of the proposed algorithm increase by about 5% and 10%, respectively, compared to the SiamRPN++ algorithm when running on the dataset of GOT-10k (Generic Object Tracking Benchmark).

Sign in / Sign up

Export Citation Format

Share Document