Hand Tracking Using Optical-Flow Embedded Particle Filter in Sign Language Scenes

Belgacem, Selma; Chatelain, Clément; Ben-Hamadou, Achraf; Paquet, Thierry

doi:10.1007/978-3-642-33564-8_35

Cited by 6 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One way of improving the estimate

would be to adapt the motion model of the particle filter. Several approaches can be found in literature that incorporate the latest measurements directly into the motion model of a particle filter [ 41 , 42 ]. However, from a Bayesian perspective, the motion model of the particle filter, yielding the state transition prior p ( x t | x t −1 ), should only depend on the previous state estimate and not on the current observations.…”

Section: Methodsmentioning

confidence: 99%

Robust Arm and Hand Tracking by Unsupervised Context Learning

Spruyt

Ledda

Philips

2014

Sensors

View full text Add to dashboard Cite

Hand tracking in video is an increasingly popular research field due to the rise of novel human-computer interaction methods. However, robust and real-time hand tracking in unconstrained environments remains a challenging task due to the high number of degrees of freedom and the non-rigid character of the human hand. In this paper, we propose an unsupervised method to automatically learn the context in which a hand is embedded. This context includes the arm and any other object that coherently moves along with the hand. We introduce two novel methods to incorporate this context information into a probabilistic tracking framework, and introduce a simple yet effective solution to estimate the position of the arm. Finally, we show that our method greatly increases robustness against occlusion and cluttered background, without degrading tracking performance if no contextual information is available. The proposed real-time algorithm is shown to outperform the current state-of-the-art by evaluating it on three publicly available video datasets. Furthermore, a novel dataset is created and made publicly available for the research community.

show abstract

“…One way of improving the estimate

Section: Methodsmentioning

confidence: 99%

Robust Arm and Hand Tracking by Unsupervised Context Learning

Spruyt

Ledda

Philips

2014

Sensors

View full text Add to dashboard Cite

show abstract

“…Shan et al [6] adopted the mean shift embedded particle filter as a non-linear posterior density estimator for real time hand tracking. Belgacem et al [7] likewise embedded optical flow as a penalisation method into particle filter for sign language recognition. Campr et al [8] used joint particle filter to calculate a combined likelihood model of hands and head.…”

Section: Related Workmentioning

confidence: 99%

A four dukkha state-space model for hand tracking

Lim

Tan

2017

Neurocomputing

View full text Add to dashboard Cite

In this paper, we propose a hand tracking method which was inspired by the notion of the four dukkha: birth, aging, sickness and death (BASD) in Buddhism. Based on this philosophy, we formalize the hand tracking problem in the BASD framework, and apply it to hand track hand gestures in isolated sign language videos. The proposed BASD method is a novel nature-inspired computational intelligence method which is able to handle complex real-world tracking problem. The proposed BASD framework operates in a manner similar to a standard state-space model, but maintains multiple hypotheses and integrates hypothesis update and propagation mechanisms that resemble the effect of BASD. The survival of the hypothesis relies upon the strength, aging and sickness of existing hypotheses, and new hypotheses are birthed by the fittest pairs of parent hypotheses. These properties resolve the sample impoverishment problem of the particle filter. The estimated hand trajectories show promising results for the American sign language.

show abstract

“…They were first introduced in the famous condensation paper of Isard and Blake [56]. They have been shown to be efficient and effective tracking methods in a variety of contexts [14,85,93] such as human, vehicle and sports tracking. They have been shown to be especially robust to complex environments.…”

Section: Particle Filtersmentioning

confidence: 99%