Recent trends in multicue based visual tracking: A review

Kumar, Ashish; Walia, Gurjit Singh; Sharma, Kapil

doi:10.1016/j.eswa.2020.113711

Cited by 45 publications

(10 citation statements)

References 131 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Walia et al [26], on the other hand, presented a multicue particle filter method based on fusion of color and texture information and a novel resampling method based upon crow search optimization to overcome low performing particles detected as the outlier. An extensive review of the recent trends in multi-cue based visual tracking can be seen in the work of Kumar et al [4].…”

Section: Background Workmentioning

confidence: 99%

Real-time Motion Tracking Enhancement via Data-Fusion based Particle Filter

2021

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Motion tracking is a well-defined yet application-specific problem of computer vision field, mostly entailing real-time constraints. Methods addressing such problems are expected also to ensure achievements such as high accuracy and robustness. A probabilistic estimation-based approach is proposed in this paper, in order to enhance the real-time motion tracking process of an RGB-Depth device, in terms of accuracy. A novel method is presented for tracking hand-palm of a moving human subject to this end, under a sequence of assumptions such as indoor environment, single object, smooth movement and stable illumination. Tracking accuracy is improved within a particle filter framework by fusing device output with the newly-extracted information from RGB and depth images. Experimental results are shared revealing the advantages of the proposed method over the built-in device algorithms. The results demonstrate that the proposed method produces smaller RMSE values both for single implementations and multi-execution trials without violating real-time constraints.

show abstract

Section: Background Workmentioning

confidence: 99%

Real-time Motion Tracking Enhancement via Data-Fusion based Particle Filter

2021

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…There are many researches to solve such kind of problems including applications related to diverse fields. Those are mainly categorized into four approaches in the field literature: stochastic, deterministic, productive, and discriminative [4]. Although each approach has distinct advantages and disadvantages over the others, there are common problematic issues that ultimately lead to the curse of dimensionality for almost all methods such as complex background, illumination changes, shape deformation, occlusion [5], abrupt and rapid movements as well as real-time constraints.…”

Section: Background Workmentioning

confidence: 99%

“…Walia et al [26], on the other hand, presented a multicue particle filter method based on fusion of color and texture information and a novel resampling method based upon crow search optimization to overcome low performing particles detected as the outlier. An extensive review of the recent trends in multicue based visual tracking can be seen in the work of Kumar et al [4].…”

Section: Background Workmentioning

confidence: 99%

Real-time motion tracking enhancement via data-fusion based particle filter

Tasci¹,

Çelebi²

2021

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Motion tracking is a well-defined yet application-specific problem of computer vision field, mostly entailing real-time constraints. Methods addressing such problems are expected also to ensure achievements such as high accuracy and robustness. A probabilistic estimation-based approach is proposed in this paper, in order to enhance the real-time motion tracking process of an RGB-Depth device, in terms of accuracy. A novel method is presented for tracking handpalm of a moving human subject to this end, under a sequence of assumptions such as indoor environment, single object, smooth movement and stable illumination. Tracking accuracy is improved within a particle filter framework by fusing device output with the newly extracted information from RGB and depth images. Experimental results are shared revealing the advantages of the proposed method over the built-in device algorithms. The results demonstrate that the proposed method produces smaller RMSE values both for single implementations and multiexecution trials without violating real-time constraints.

show abstract

“…Over many years, machine learning (ML) algorithms have been used to extract different types of features such as color, texture, shape, thermal, and audio. Researchers have mainly concentrated on integrating two or more ML algorithms to cover most features and achieve good tracking results [6]. In the last few years, researchers have shifted their procedure to deep learning (DL) algorithms, which have achieved remarkable performance compared to ML algorithms.…”

Section: Introductionmentioning

confidence: 99%

Robust visual tracking using very deep generative model

AlBasiouny

Attia²,

Abdelmunim³

et al. 2023

J Big Data

View full text Add to dashboard Cite

Deep learning algorithms provide visual tracking robustness at an unprecedented level, but realizing an acceptable performance is still challenging because of the natural continuous changes in the features of foreground and background objects over videos. One of the factors that most affects the robustness of tracking algorithms is the choice of network architecture parameters, especially the depth. A robust visual tracking model using a very deep generator (RTDG) was proposed in this study. We constructed our model on an ordinary convolutional neural network (CNN), which consists of feature extraction and binary classifier networks. We integrated a generative adversarial network (GAN) into the CNN to enhance the tracking results through an adversarial learning process performed during the training phase. We used the discriminator as a classifier and the generator as a store that produces unlabeled feature-level data with different appearances by applying masks to the extracted features. In this study, we investigated the role of increasing the number of fully connected (FC) layers in adversarial generative networks and their impact on robustness. We used a very deep FC network with 22 layers as a high-performance generator for the first time. This generator is used via adversarial learning to augment the positive samples to reduce the gap between the hungry deep learning algorithm and the available training data to achieve robust visual tracking. The experiments showed that the proposed framework performed well against state-of-the-art trackers on OTB-100, VOT2019, LaSOT and UAVDT benchmark datasets.

show abstract

Recent trends in multicue based visual tracking: A review

Cited by 45 publications

References 131 publications

Real-time Motion Tracking Enhancement via Data-Fusion based Particle Filter

Real-time Motion Tracking Enhancement via Data-Fusion based Particle Filter

Real-time motion tracking enhancement via data-fusion based particle filter

Robust visual tracking using very deep generative model

Contact Info

Product

Resources

About