Dynamic Codebook for Foreground Segmentation in a Video

The foreground segmentation in a video is a way to extract changes in image sequences. It is a key task in an early stage of many applications in the computer vision area. The information of changes in the scene must be segmented before any further analysis could be taken place. However, it remains with difficulties caused by several real-world challenges such as cluttered backgrounds, changes of the illumination, shadows, and long-term scene changes. This paper proposes a novel method, namely a dynamic codebook (DCB), to address such challenges of the dynamic backgrounds. It relies on a dynamic modeling of the background scene. Initially, a codebook is constructed to represent the background information of each pixel over a period of time. Then, a dynamic boundary of the codebook will be made to support variations of the background. The revised codebook will always be adaptive to the new background's environments. This makes the foreground segmentation more robust to the changes of background scene. The proposed method has been evaluated by using the changedetection.net (CDnet) benchmark which is a well-known video dataset for testing change-detection algorithms. The experimental results and comprehensive comparisons have shown a very promising performance of the proposed method.

[1]  Larry S. Davis,et al.  A Robust Background Subtraction and Shadow Detection , 1999 .

[2]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[3]  Olivier Bernier,et al.  Local kernel color histograms for background subtraction , 2006, VISAPP.

[4]  Alex Pentland,et al.  Pfinder: real-time tracking of the human body , 1996, Proceedings of the Second International Conference on Automatic Face and Gesture Recognition.

[5]  Hélène Laurent,et al.  Comparative study of background subtraction algorithms , 2010, J. Electronic Imaging.

[6]  Worapan Kusakunniran,et al.  Recognizing Gaits on Spatio-Temporal Feature Domain , 2014, IEEE Transactions on Information Forensics and Security.

[7]  Bin Wang,et al.  A Fast Self-Tuning Background Subtraction Algorithm , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[8]  Hasan Sajid,et al.  Universal Multimode Background Subtraction , 2017, IEEE Transactions on Image Processing.

[9]  Xiqun Lu,et al.  A multiscale spatio-temporal background model for motion detection , 2014, 2014 IEEE International Conference on Image Processing (ICIP).

[10]  Ferdinand van der Heijden,et al.  Efficient adaptive density estimation per image pixel for the task of background subtraction , 2006, Pattern Recognit. Lett..

[11]  Cina Motamed,et al.  Foreground-Background Segmentation Based on Codebook and Edge Detector , 2014, 2014 Tenth International Conference on Signal-Image Technology and Internet-Based Systems.

[12]  William J. Christmas,et al.  A Tennis Ball Tracking Algorithm for Automatic Annotation of Tennis Match , 2005, BMVC.

[13]  D. Koller,et al.  Towards robust automatic traffic scene analysis in real-time , 1994, Proceedings of 1994 33rd IEEE Conference on Decision and Control.

[14]  S. Bianco,et al.  How Far Can You Get By Combining Change Detection Algorithms? , 2015, ICIAP.

[15]  Serge Miguet,et al.  Real Time Foreground-Background Segmentation Using a Modified Codebook Model , 2009, 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance.

[16]  Fatih Murat Porikli,et al.  CDnet 2014: An Expanded Change Detection Benchmark Dataset , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[17]  Hasan Sajid,et al.  Background subtraction for static & moving camera , 2015, 2015 IEEE International Conference on Image Processing (ICIP).

[18]  Rui Wang,et al.  Static and Moving Object Detection Using Flux Tensor with Split Gaussian Models , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[19]  Wilfried Philips,et al.  EFIC: Edge Based Foreground Background Segmentation and Interior Classification for Dynamic Camera Viewpoints , 2015, ACIVS.

[20]  Anurag Singh,et al.  Object Tracking Using Frame Differencing and Template Matching , 2012 .

[21]  W. Eric L. Grimson,et al.  Adaptive background mixture models for real-time tracking , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[22]  Atta Badii,et al.  Change detection based on graph cuts , 2015, 2015 International Conference on Systems, Signals and Image Processing (IWSSIP).

[23]  Ezequiel López-Rubio,et al.  Color space selection for self-organizing map based foreground detection in video sequences , 2014, 2014 International Joint Conference on Neural Networks (IJCNN).

[24]  Andrew K. C. Wong,et al.  A new method for gray-level picture thresholding using the entropy of the histogram , 1985, Comput. Vis. Graph. Image Process..

[25]  Mario I. Chacon-Murguia,et al.  Auto-Adaptive Parallel SOM Architecture with a modular analysis for dynamic object segmentation in videos , 2016 .

[26]  Benjamin Höferlin,et al.  Evaluation of background subtraction techniques for video surveillance , 2011, CVPR 2011.

[27]  Wilfried Philips,et al.  C-EFIC: Color and Edge Based Foreground Background Segmentation with Interior Classification , 2015, VISIGRAPP.

[28]  Mansour Moniri,et al.  Spectral-360: A Physics-Based Technique for Change Detection , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[29]  Chung-Lin Huang,et al.  Hybrid Codebook Model for Foreground Object Segmentation and Shadow/Highlight Removal , 2014, J. Inf. Sci. Eng..

[30]  Massimo De Gregorio,et al.  Change Detection with Weightless Neural Networks , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[31]  Zoran Zivkovic,et al.  Improved adaptive Gaussian mixture model for background subtraction , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[32]  Huiyu Zhou,et al.  Spatial mixture of Gaussians for dynamic background modelling , 2013, 2013 10th IEEE International Conference on Advanced Video and Signal Based Surveillance.

[33]  Lucia Maddalena,et al.  A Self-Organizing Approach to Background Subtraction for Visual Surveillance Applications , 2008, IEEE Transactions on Image Processing.

[34]  Mario Ignacio Chacon Murguia,et al.  Auto-Adaptive Parallel SOM Architecture with a modular analysis for dynamic object segmentation in videos , 2016, Neurocomputing.

[35]  Dong Liang,et al.  Improvements and Experiments of a Compact Statistical Background Model , 2014, ArXiv.

[36]  Massimo De Gregorio,et al.  WiSARDrp for Change Detection in Video Sequences , 2017, ESANN.

[37]  Youtian Du,et al.  Gaussian-Based Codebook Model for Video Background Subtraction , 2006, ICNC.

[38]  Larry S. Davis,et al.  Real-time foreground-background segmentation using codebook model , 2005, Real Time Imaging.

[39]  Larry S. Davis,et al.  Non-parametric Model for Background Subtraction , 2000, ECCV.

[40]  Ronald Poppe,et al.  A survey on vision-based human action recognition , 2010, Image Vis. Comput..

[41]  Lucia Maddalena,et al.  A fuzzy spatial coherence-based approach to background/foreground separation for moving object detection , 2010, Neural Computing and Applications.

[42]  Ghazali Sulong,et al.  Vehicle Detection and Tracking Techniques: A Concise Review , 2014, ArXiv.