Skeleton-Based Human Action Recognition on Large-Scale Datasets

Hossain, Tonmoy; Sarker, Sujan; Rahman, Sejuti; Ahad, Md. Atiqur Rahman

doi:10.1007/978-3-030-75490-7_5

Cited by 6 publications

(4 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these data have problems such as a large amount of information to process, high amount of noise, difficulty in recording data, and high cost [13]. 3D skeleton-based data are widely used in recent research to mitigate some of the problems of the other data types [2], [3], [4], [5], [6]. Most methods introduced for HAR are supervised focusing in Fig 1 part (d) and (e), training from labeled data.…”

Section: Human Activity Recognition and Discoverymentioning

confidence: 99%

See 1 more Smart Citation

Flexible multi-objective particle swarm optimization clustering with game theory to address human activity discovery fully unsupervised

Hadikhani,

Lai,

Ong

2024

Image and Vision Computing

View full text Add to dashboard Cite

Section: Human Activity Recognition and Discoverymentioning

confidence: 99%

“…E-mail: {20h8561, daphne.lai, weehong.ong}@ubd.edu.bn. [4], [5], [6]. Further, depth sensor have the advantage of not capturing personal identity images.…”

Section: Introductionmentioning

confidence: 99%

Flexible multi-objective particle swarm optimization clustering with game theory to address human activity discovery fully unsupervised

Hadikhani,

Lai,

Ong

2024

Image and Vision Computing

View full text Add to dashboard Cite

“…In most cases, the dataset is large enough to experiment with deep learning approaches. Deep learning methods are expected to outperform classical machine learning models with hand-crafted features [10]. In this case, a large dataset is advantageous for a data-driven method.…”

Section: Related Workmentioning

confidence: 99%

Can Ensemble of Classifiers Provide Better Recognition Results in Packaging Activity?

Sakib

Basak

Uddin

et al. 2022

Sensor- And Video-Based Activity and Behavior Computing

Self Cite

View full text Add to dashboard Cite

Skeleton-based Motion Capture (MoCap) systems have been widely used in the game and film industry for mimicking complex human actions for a long time. MoCap data has also proved its effectiveness in human activity recognition tasks. However, it is a quite challenging task for smaller datasets. The lack of such data for industrial activities further adds to the difficulties. In this work, we have proposed an ensemble-based machine learning methodology that is targeted to work better on MoCap datasets. The experiments have been performed on the MoCap data given in the Bento Packaging Activity Recognition Challenge 2021. Bento is a Japanese word that resembles lunch-box. Upon processing the raw MoCap data at first, we have achieved an astonishing accuracy of 98% on 10-fold Cross-Validation and 82% on Leave-One-Out-Cross-Validation by using the proposed ensemble model.

show abstract

“…In the first step (Fig 1(a)), people's activities are recorded. Recent researches have shown that 3D-skeleton data is reliable and can be easily recorded with low-cost depth sensors [2], [3], [4], [5], [6]. Further, depth sensor have the advantage of not capturing personal identity images.…”

Section: Introductionmentioning

confidence: 99%

Flexible Multi-Objective Particle Swarm Optimization Clustering with Game Theory to Address Human Activity Discovery Fully Unsupervised

Hadikhani¹,

Lai²,

Ong³

2022

Preprint

View full text Add to dashboard Cite

<p>Most research in human activity recognition is supervised, while non-supervised approaches are not completely unsupervised. In this paper, we provide a novel flexible multi-objective particle swarm optimization (PSO) clustering method based on game theory (FMOPG) to discover human activities fully unsupervised. Unlike conventional clustering methods that estimate the number of clusters and are very time-consuming and inaccurate, an incremental technique is introduced which makes the proposed method flexible in dealing with the number of clusters. Using this technique, clusters that have a better connectedness and good separation from other clusters are gradually selected. To improve the convergence speed of PSO in achieving the best solution and dealing with non spherical shape clusters, updating of particles' velocity is modified using the concept of mean-shift vector. To solve multi-objective optimization problems, Nash equilibrium in game theory is used to select the optimal solution on the pareto front. Gaussian mutation is also employed on the pareto front to generate diverse solutions and create a balance between exploitation and exploration. The proposed method is compared with state-of-the-art methods on seven challenging datasets. FMOPG has improved clustering accuracy by 3.65% compared to other automated methods. Moreover, the incremental technique has improved the clustering time by 71.18%.</p>

show abstract

Skeleton-Based Human Action Recognition on Large-Scale Datasets

Cited by 6 publications

References 66 publications

Flexible multi-objective particle swarm optimization clustering with game theory to address human activity discovery fully unsupervised

Flexible multi-objective particle swarm optimization clustering with game theory to address human activity discovery fully unsupervised

Can Ensemble of Classifiers Provide Better Recognition Results in Packaging Activity?

Flexible Multi-Objective Particle Swarm Optimization Clustering with Game Theory to Address Human Activity Discovery Fully Unsupervised

Contact Info

Product

Resources

About