Abstract An algorithm is proposed to segment foreground accurately from videos whose background is dynamic or whose foreground performs non-translational motion. Firstly, by regarding the change process of a single pixel as discrete-time signal, the video is segmented into foreground and background in a glancing way with temporal analysis using Gabor filter. Secondly, global color model and local color model are defined and built by clustering the color information of the background and foreground with mean-shift algorithm. Finally, a double-labeling method is used for fine segmentation of the foreground. Experimental results on several datasets prove that the proposed algorithm evidently improves the precision of the extracted foreground, especially in the cases that the background is dynamic or the foreground performs non-translational motion.
[1] Stauffer C,Grimson W E L.Adaptive Background Mixture Models for Real-Time Tracking // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Fort Collins,USA,1999,2: 246-252 [2] Kaewtrakulpong P,Bowden R.An Improved Adaptive Background Mixture Model for Real-Time Tracking with Shadow Detection // Proc of the 2nd European Workshop on Advanced Video-Based Surveillance Systems.Kingston,UK,2001: 149-158 [3] Elgammal A,Duraiswami R,Harwood D,et al.Background and Foreground Modeling Using Nonparametric Kernel Density Estimation for Visual Surveillance // Proc of the IEEE,2002,90(7): 1151-1163 [4] Migdal J,Grimson E L W.Background Subtraction Using Markov Thresholds // Proc of the IEEE Workshop on Motion and Video Computing.Washington D C,USA,2005: 58-65 [5] Sheikh Y,Shah M.Bayesian Modeling of Dynamic Scenes for Object Detection.IEEE Trans on Pattern Analysis and Machine Intelligence,2005,27(11): 1778-1792 [6] Beigi H S M,Maes S H,Maes S H.A Hierarchical Approach to Large-Scale Speaker Recognition // Proc of the 6th European Conference on Speech Communication and Technology.Budapest,Hungary,1999,5: 2203-2206 [7] Chen Cheng,Zhuang Yueting,Xiao Jun.Video Foreground Segmentation with Camera Movement.Journal of Zhejiang University (Engineering Science),2009,43(6): 973-977 (in Chinese) (陈 成,庄越挺,肖 俊.相机运动条件下的视频前景提取.浙江大学学报(工学版),2009,43(6): 973-977) [8] Fukunaga K,Hostetler L.The Estimation of the Gradient of a Density Function with Applications in Pattern Recognition.IEEE Trans on Information Theory,1975,21(1): 32-40 [9] Cheng Yizong.Mean Shift,Mode Seeking,and Clustering.IEEE Trans on Pattern Analysis and Machine Intelligence,1995,17(8): 790-799 [10] Comaniciu D,Meer P.Mean Shift: A Robust Approach toward Feature Space Analysis.IEEE Trans on Pattern Analysis and Machine Intelligence,2002,24(5): 603-619 [11] Gorelick L,Blank M,Shechtman E,et al.Actions as Space-Time Shapes.IEEE Trans on Pattern Analysis and Machine Intelligence,2005,29(12): 2247-2253 [12] Dollar P,Rabaud V,Cottrell G,et al.Behavior Recognition via Sparse Spatio-Temporal Features // Proc of the 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Survillance.Beijing,China,2005: 65-72 [13] Laptev I,Lindeberg T.Space-Time Interest Points // Proc of the 9th IEEE International Conference on Computer Vision.Nice,France,2003,Ⅰ: 432-439
2011, Vol. 24 
