IMBS Background Subtraction Library

Domenico Daniele Bloisi


Independent Multimodal Background Subtraction (IMBS) Library is a C++ library designed for performing an accurate foreground extraction in real-time. IMBS creates a multimodal model of the background in order to deal with illumination changes, camera jitter, movements of small background elements, and changes in the background geometry. A statistical analysis of the frames in input is performed to obtain the background model. Bootstrapis requiredin order tobuild theinitial background model.IMBS exploits OpenCV functions.



Shadow detection on video sequences from the BMC data set.


Documentation

A detailed description of the IMBS method can be found in the paper "Independent Multimodal Background Subtraction" (draft) in the proceedings of Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications (CompImage) 2012

Please, cite the above paper if you use IMBS.


Get IMBS

IMBS is provided without any warranty about its usability. It is for educational purposes and should be regarded as such.

  • IMBS 3.1 (imbs version compatible with OpenCV 3 with bug fixes)

    Source code for IMBS 3.1 (can be compiled with cmake) can be downloaded here.

    IMBS 3.1 example on a video sequence from PETS 2009.

  • IMBS 3 (imbs version compatible with OpenCV 3)

    Source code for IMBS 3 (can be compiled with cmake) can be downloaded here.

    IMBS 3 example on a video sequence from PETS 2009.

  • FAFEX - Fast Adaptive Foreground Extraction (multi-threaded IMBS)

    Source code for FAFEX (can be compiled with cmake) can be downloaded here.

    FAFEX example on a video sequence from the BMC data set.

  • IMBS 2 (imbs version compatible with OpenCV 2)

    Source code for IMBS 2 (can be compiled with cmake) can be downloaded here.

  • IMBS 1 (imbs version compatible with OpenCV 1)

    The old page for IMBS 1 is here.


IMBS Prerequisites


BuildIMBS

IMBS is provided with a CMakeLists.txt file and can be compiled by using CMake.

Linux

  1. Unzip the file imbs3.zip in
  2. $cd
  3. $mkdir build
  4. $cd build
  5. $cmake ..
  6. $make

Windows

  1. Unzip the file imbs3.zip
  2. Use the CMake graphical user interface to create the desired makefile

RunIMBS

IMBS is provided with an usage example (main.cpp)

Linux

For video files

$./imbs -vid video1.avi

For an image sequence (fps = 25 default value)

$./imbs -img images/1.png

or you can specify the fps value

$./imbs -img images/1.png -fps 7

Note that the usage example works only on image sequences in which the filename format is .png, where n is the frame number (e.g., 7.png).

Windows

For video files

>imbs -vid video1.avi

For an image sequence (fps = 25 default value)

>imbs -img images/1.png

or you can specify the fps value

>imbs -img images/1.png -fps 7

Note: For IMBS 2 the usage example works only on image sequences in which the filename format is .png, where n is the frame number (e.g., 7.png).


Results

IMBS has been tested with multiple well-known image sequences containing dynamic background. Results obtained by using IMBS without applying any morphological operator are reported in the following.

  1. Waving trees [1]    raw data (4 fps)    imbs results
  2. Canoe [2]    raw data (25 fps)    imbs results


Publications

[1] D.D. Bloisi, A. Pennisi, L. Iocchi, "Background modeling in the maritime domain", In Machine Vision and Applications, Springer Berlin Heidelberg, vol. 25, no. 5, pp. 1257-1269, 2014. draft

[2] D.D. Bloisi, "Background Modeling and Foreground Detection for Maritime Video Surveillance", Chapter in Handbook on Background Modeling and Foreground Detection for Video Surveillance: Traditional and Recent Approaches, Implementations, Benchmarking and Evaluation, Chapman and Hall/CRC, pp. 14-1-14-22, 2014. draft

[3] D.D. Bloisi, L. Iocchi, "Independent Multimodal Background Subtraction", In Proceedings of the Third International Conference on Computational Modeling of Objects Presented in Images: Fundamentals, Methods and Applications, Rome, Italy, pp. 39-44, 2012. draft


About

IMBS has been written by Domenico Daniele Bloisi.

The author thanks Loris Bazzani, Andrea Pennisi, and Fabio Previtali.


References

[1] Kentaro Toyama, John Krumm, Barry Brumitt, Brian Meyers, "Wallflower: Principles and Practice of Background Maintenance", Seventh International Conference on Computer Vision, September 1999, Kerkyra, Greece, pp. 255-261, IEEE Computer Society Press.

[2] N. Goyette, P.-M. Jodoin, F. Porikli, J. Konrad, and P. Ishwar, changedetection.net: A new change detection benchmark dataset, in Proc. IEEE Workshop on Change Detection (CDW12) at CVPR12, Providence, RI, 16-21 Jun., 2012.