Abstract:A key problem in automated surveillance systems is to detect foreground accurately in image sequence. However, it is difficult in complex scenes. In this paper, we consider the problem as a labeling problem and present a multi-scale discriminative model to learn complex background for foreground detection. First, the static background is obtained by pixel-wise methods. At the same time, the periodic motions, such as swaying tree, water wave and moving shadow, will be wrongly detected as foreground, due to they have the same motion feature with true foreground. The detected results in this step are denoted as moving objects. Second, the pixels in the moving objects can be classified as dynamic background and foreground associated with the confidence by a boosted classifier. A Gaussian filter bank with different variances is exploited to form multi-scale images in different image spaces at the beginning, then a feature pool is obtained by kernel density estimation on the image sequence over time. The boosted classifier is trained by AdaBoost over the feature pool and the labeled positive and negative data. Third, Markov random field (MRF) model is used to infer the spatial and temporal coherence over the labels for foreground/background segmentation accurately. Experiments tested on the various videos show that the proposed method can be work well on the complex background.

A multimodal background model based on binning kernel density estimation

Background modeling using mixture of Gaussians and Laplacian pyramid decomposition.

A Multimodal Nonparametric Background Model for Moving Object Detection

Tracking human under occlusion based on kernel density estimation

Gaussian Kernel Density Estimation-based Background Modeling with Noise and Shadow Suppression

Vision-Based Moving Objects Detection with Background Modeling

Multimodal background model with noise and shadow suppression for moving object detection

Learning Complex Background Based on Multi-scale Discriminative Model

Improved kernel density background estimation with diversity sampling and neighbor information for traffic monitoring

Diversity Sampling Based Kernel Density Estimation for Background Modeling

Learning complex background by multi-scale discriminative model

Multi-Channel Adaptive Mixture Background Model for Real-time Tracking.

Kernel Density Estimation With Diversity Samples For Multimodal Scene Modeling

A temporal-spatial background modeling of dynamic scenes

A multiscale spatio-temporal background model for motion detection

Background extraction model and shadow detection algorithm in traffic flow video detection

Background Modeling Based On Region Segmentation

Pedestrian detection based on modified dynamic background using gaussian mixture models and hog-svm detection

A background model estimation algorithm based on analysis of local motion for video surveillance

Moving area detection based on estimation of static background

A General Moving Detection Method Using Dual-Target Nonparametric Background Model.