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 Multi-Histogram Clustering Approach Toward Markov Random Field for Foreground Segmentation.

Adaptive Foreground and Shadow Segmentation Using Hidden Conditional Random Fields

Vibration-resisted video segmentation algorithm based on hierarchical Markov random field model

Segmentation Algorithm of Moving Object Based on Mean Field Theory and MRF-MAP

Background Segmentation Using Spatial-Temporal Multi-Resolution MRF.

Video Foreground Segmentation Based on Sequential Feature Clustering

Monocular Video Foreground/Background Segmentation by Tracking Spatial-Color Gaussian Mixture Models

A Bayesian Approach to Clustering Matting Components in Spectral Matting

A Multi-Layer MRF Model Fusing Entropy Information for Foreground Segmentation in Video Sequences

Learning complex background by multi-scale discriminative model

Learning Complex Background Based on Multi-scale Discriminative Model

Dynamic Background Subtraction Using Histograms Based on Fuzzy C-Means Clustering and Fuzzy Nearness Degree.

Adaptive Fuzzy Clustering Algorithm With Local Information And Markov Random Field For Image Segmentation

Traffic Video Image Segmentation Model Based on Bayesian and Spatio-Temporal Markov Random Field

Adaptive Foreground and Shadow Detection inImage Sequences

Efficient video object segmentation based on Gaussian mixture model and Markov random field

A New Multi-threshold Segmentation Method Based on MHFFCM

An Effective Method for Foreground Segmentation of Video

A Threshold Segmentation Method Based On Fuzzy C-Means Clustering Algorithm And Multi-Histogram

Video object segmentation by motion-based sequential feature clustering.

A Robust Foreground Segmentation Method by Temporal Averaging Multiple Video Frames