Moving object detection for real-time augmented reality applications in a GPGPU

Cuevas, Carlos; Berjón, Daniel; Morán, F.; García, Narciso

doi:10.1109/tce.2012.6170063

Cited by 21 publications

(16 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…l.c and Fig. l.d present the detections resulting from the combination of the background modeling in [3] and the foreground modeling in [4], using the typical Bayesian classifier [6] (Fig. l.c), and using the one we propose (Fig.…”

Section: Spatially Conditioned Nonparametric Modelingmentioning

confidence: 99%

“…This drawback can be solved by combining the background model in [3] with the foreground model in [4], but then two new issues appear. Since the foreground modeling requires kernels with larger spatial widths [4], the spatial density of the foreground model is much higher than the spatial density of the background, which may lead to false detections (see Fig. l.c).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Region-based moving object detection using spatially conditioned nonparametric models in a GPU

Berjón

Cuevas

Morán

et al. 2014

2014 IEEE International Conference on Consumer Electronics (ICCE)

Self Cite

View full text Add to dashboard Cite

Abstract-A novel GPU-based nonparametric moving object detection strategy for computer vision tools requiring real-time processing is proposed. An alternative and efficient Bayesian classifier to combine nonparametric background and foreground models allows increasing correct detections while avoiding false detections. Additionally, an efficient region of interest analysis significantly reduces the computational cost of the detections.

show abstract

Section: Spatially Conditioned Nonparametric Modelingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Region-based moving object detection using spatially conditioned nonparametric models in a GPU

Berjón

Cuevas

Morán

et al. 2014

2014 IEEE International Conference on Consumer Electronics (ICCE)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Observe that most cited works have dimension n = 2, 3 or 5 since they correspond to models that use spatial coordinates and/or the color vector of a pixel as components to apply a distance criterion. The last two columns report the actual confidence value (P ) obtained by us resulting from the parameters of the model (information in columns [3][4][5]. If the criterion (column 3) is clear, only the corresponding column (Rect.…”

Section: Comparison Of Classification Criteriamentioning

confidence: 99%

“…Often, the algorithms proposed in these works require the computation of cumulative probabilities of the normal distributions for different purposes such as, for example, disregarding the data that does not contribute significantly to the distributions [5] or evaluating how well a normal distribution represents a data set [6].…”

Section: Introductionmentioning

confidence: 99%

On the Mahalanobis Distance Classification Criterion for Multidimensional Normal Distributions

Gallego

Cuevas

Mohedano

et al. 2013

IEEE Trans. Signal Process.

Self Cite

View full text Add to dashboard Cite

Abstract-Many existing engineering works model the statistical characteristics of the entities under study as normal distributions. These models are eventually used for decision making, requiring in practice the definition of the classification region corresponding to the desired confidence level. Surprisingly enough, however, a great amount of computer vision works using multidimensional normal models leave unspecified or fail to establish correct confidence regions due to misconceptions on the features of Gaussian functions or to wrong analogies with the unidimensional case. The resulting regions incur in deviations that can be unacceptable in high-dimensional models.Here we provide a comprehensive derivation of the optimal confidence regions for multivariate normal distributions of arbitrary dimensionality. To this end, firstly we derive the condition for region optimality of general continuous multidimensional distributions, and then we apply it to the widespread case of the normal probability density function. The obtained results are used to analyze the confidence error incurred by previous works related to vision research, showing that deviations caused by wrong regions may turn into unacceptable as dimensionality increases. To support the theoretical analysis, a quantitative example in the context of moving object detection by means of background modeling is given.

show abstract

“…Furthermore, to improve the quality of the detections in scenarios where moving objects and background have similar characteristics [10], most recent nonparametric proposals estimate not only a background density function but also a foreground model, combining both models by means of a Bayesian classifier [11]. However, typically used Bayesian classifiers only use prior probabilities defined constantly over the whole image extent [12].…”

mentioning

confidence: 99%

Adaptable Bayesian classifier for spatiotemporal nonparametric moving object detection strategies

2012

Self Cite

View full text Add to dashboard Cite

Electronic devices endowed with camera platforms require new and powerful machine vision applications, which commonly include moving object detection strategies. To obtain high-quality results, the most recent strategies estimate nonparametrically background and foreground models and combine them by means of a Bayesian classifier. However, typical classifiers are limited by the use of constant prior values and they do not allow the inclusion of additional spatiodependent prior information. In this Letter, we propose an alternative Bayesian classifier that, unlike those reported before, allows the use of additional prior information obtained from any source and depending on the spatial position of each pixel.

show abstract

Moving object detection for real-time augmented reality applications in a GPGPU

Cited by 21 publications

References 31 publications

Region-based moving object detection using spatially conditioned nonparametric models in a GPU

Region-based moving object detection using spatially conditioned nonparametric models in a GPU

On the Mahalanobis Distance Classification Criterion for Multidimensional Normal Distributions

Adaptable Bayesian classifier for spatiotemporal nonparametric moving object detection strategies

Contact Info

Product

Resources

About