Bayesian modeling of dynamic scenes for object detection

Abstract

Accurate detection of moving objects is an important precursor to stable tracking or recognition. In this paper, we present an object detection scheme that has three innovations over existing approaches.Firstly, the model of the intensities of image pixels as independent random variables is challenged and it is asserted that useful correlation exists in intensities of spatially proximal pixels. This correlation is exploited to sustain high levels of detection accuracy in the presence of dynamic backgrounds. By using a non-parametric density estimation method over a joint domain-range representation of image pixels, multi-modal spatial uncertainties and complex dependencies between the domain (location) and range (color) are directly modeled. We propose a model of the background as a single probability density.Secondly, temporal persistence is proposed as a detection criterion. Unlike previous approaches to object detection which detect objects by building adaptive models of the background, the foreground is modeled to augment the detection of objects (without explicit tracking), since objects detected in the preceding frame contain substantial evidence for detection in the current frame. Finally, the background and foreground models are used competitively in a MAP-MRF decision framework, stressing spatial context as a condition of detecting interesting objects and the posterior function is maximized efficiently by finding the minimum cut of a capacitated graph. Experimental validation of the proposed method is performed and presented on a diverse set of dynamic scenes.

Making Background Subtraction Robust to Sudden Illumination Changes

“…Elgammal et al proposed to model both background and foreground pixel intensities by a nonparametric kernel density estimation [6]. In [7], Sheikh and Shah proposed to model the full background with a single distribution, instead of one distribution per pixel, and to include location into the model. Because these methods do not decouple illumination from other causes of background changes, they are more sensitive to drastic light effects than our approach.…”

Background foreground segmentation with RGB-D Kinect data: An efficient combination of classifiers

“…In the proposed approach we use a uniform distribution to model the foreground process (as proposed in [27]): p(x s |c%) = J. w .…”

Figure-Ground Segmentation—Pixel-Based

“…Such layered representation facilitates detecting the foreground under static or dynamic background and in the presence of nominal camera motion. In [49] KDE was used in a joint domain-range representation of image pixel (r,g,b,x,y), which exploits the spatial correlation between neighboring pixels. Parag et al [41] proposed an approach for feature selection for the KDE framework where boosting based ensemble learning was used to combine different features.…”