Real-time visual tracking with ELM augmented adaptive correlation filter

Qu, Lei; Liu, Kuixiang; Yao, Baochen; Tang, Jun; Zhang, Wei

doi:10.1016/j.patrec.2018.09.015

Cited by 7 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are other CF tackers that address many model issues such as unwanted boundary and model degradation (DDCF [29]), scale change of target objects (SAMF [30]), and drifting and long-tern failure (ELMACF [31]). With the increased popularity of attention paid to mechanisms, trackers such as AFCN [32] utilized the dynamic properties of targets for the best performance.…”

Section: 𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 ∶ 𝑇𝑃 𝑇𝑃 𝐹𝑃mentioning

confidence: 99%

Critical Overview of Visual Tracking with Kernel Correlation Filter

Yadav

Payandeh

2021

Technologies

View full text Add to dashboard Cite

With the development of new methodologies for faster training on datasets, there is a need to provide an in-depth explanation of the workings of such methods. This paper attempts to provide an understanding for one such correlation filter-based tracking technology, Kernelized Correlation Filter (KCF), which uses implicit properties of tracked images (circulant matrices) for training and tracking in real-time. It is unlike deep learning, which is data intensive. KCF uses implicit dynamic properties of the scene and movements of image patches to form an efficient representation based on the circulant structure for further processing, using properties such as diagonalizing in the Fourier domain. The computational efficiency of KCF, which makes it ideal for low-power heterogeneous computational processing technologies, lies in its ability to compute data in high-dimensional feature space without explicitly invoking the computation on this space. Despite its strong practical potential in visual tracking, there is a need for an in-depth critical understanding of the method and its performance, which this paper aims to provide. Here we present a survey of KCF and its method along with an experimental study that highlights its novel approach and some of the future challenges associated with this method through observations on standard performance metrics in an effort to make the algorithm easy to investigate. It further compares the method against the current public benchmarks such as SOTA on OTB-50, VOT-2015, and VOT-2019. We observe that KCF is a simple-to-understand tracking algorithm that does well on popular benchmarks and has potential for further improvement. The paper aims to provide researchers a base for understanding and comparing KCF with other tracking technologies to explore the possibility of an improved KCF tracker.

show abstract

Section: 𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 ∶ 𝑇𝑃 𝑇𝑃 𝐹𝑃mentioning

confidence: 99%

Critical Overview of Visual Tracking with Kernel Correlation Filter

Yadav

Payandeh

2021

Technologies

View full text Add to dashboard Cite

show abstract

“…Wang et al [32] proposed the point-to-set distance metric learning which was conducted on convolutional neural network features of the training data extracted from the starting frames. Lei Qu et al [36] integrated fast histogram of oriented gradient (FHOG) and discriminative color descriptors (DD) to further boost the tracking performance.…”

Section: The Appearance Representation In Trackingmentioning

confidence: 99%

Online non-negative discriminative dictionary learning for tracking

Wang

Yang

Zhang

2019

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

In this paper, online non-negative discriminative dictionary learning for tracking is proposed, which combines the advantages of the global dictionary learning model and the class-specific dictionary learning model. The previous algorithm based on general dictionary learning does not take into account the inter-class relations between classes and make full use of tag information. In order to improve the classification ability of dictionaries, the class correlation was proposed to guide the learning of discriminant dictionaries, which makes full use of the correlation and difference between the atomic classes of dictionaries and introduces the tag information of the categories to improve the discriminant ability of dictionaries. For this purpose, the Huber loss function and the Fisher weight coefficient is used in the discriminative term to improve computational efficiency. In addition, non-negative constraints is added on dictionaries to enhance the performance. The OTB50 and OTB100 datasets are used to evaluate our tracker and compare with related algorithm. The experimental results show that our method performs much better than the tracking method compared in this paper.

show abstract

“…Moreover, the Extreme Learning Machine (ELM) has its advantage and can be factored as a complement. For example, for handling the drift originated in severe occlusion, an online classifier based on ELM for reinitializing the object on the condition of tracking failure was used by Qu et al (2019). Bi et al (2022) proposed the ELM assisted particle filter for INS/UWB integrated quadrotor positioning.…”

Section: Introductionmentioning

confidence: 99%

A FIR filter assisted with the predictive model and ELM integrated for UWB-based quadrotor aircraft localization

Dong

et al. 2023

Satell Navig

View full text Add to dashboard Cite

To improve the accuracy of the Ultra-Wide Band (UWB) based quadrotor aircraft localization, a Finite Impulse Response (FIR) filter aided with an integration of the predictive model and Extreme Learning Machine (ELM) is proposed in this work. The FIR filter estimates the quad-rotor aircraft ’s position by fusing the positions measured with the UWB and Inertial Navigation System respectively. When the UWB dada are unavailable, both the ELM and the predictive model are used to provide the measurements, replacing those unavailable UWB data, for the FIR filter. The ELM estimates the measurement via the mapping between the one step prediction of state vector and the measurement built when the UWB data are available. For the predictive model, we mathematically describe the missing UWB data. Then, both the measurements estimated with the ELM and predictive model are employed to estimate the observations via Mahalanobis distance. The test results show that the FIR filter aided by the predictive model/ELM integrated can reduce the Cumulative Distribution Function and position Root Mean Square Error effectively when the UWB is unavailable. Compared with the ELM assisted FIR filter, the proposed FIR filter can reduce the localization error by about 48.59 %, meanwhile, the integrated method is close to the method with a better solution. Graphical Abstract

show abstract

Real-time visual tracking with ELM augmented adaptive correlation filter

Cited by 7 publications

References 13 publications

Critical Overview of Visual Tracking with Kernel Correlation Filter

Critical Overview of Visual Tracking with Kernel Correlation Filter

Online non-negative discriminative dictionary learning for tracking

A FIR filter assisted with the predictive model and ELM integrated for UWB-based quadrotor aircraft localization

Contact Info

Product

Resources

About