Moving Object Detection

Objectives

The activity concerns the analysis, design, and implementation of machine learning methods for the detection, tracking and real-time recognition of objects in motion sequences, with the principal aim of developing robust and accurate methods that are suitable for real-time applications.

Main achievements

Self-organizing approach to background subtraction through artificial neural networks

- SOBS: the 2D neural model [2]. See here the SOBS software and here results on the changedetection.net dataset
- SOBS and tracking [3]
- SOBS_CF: handling uncertainty [4]. See here results on the changedetection.net dataset
- SC-SOBS: exploiting spatial coherence [8]. See here the SC-SOBS software and here results on the changedetection.net dataset
- 3D-SOBS: the 3D neural model [9]
- PTZ-SOBS: SOBS for PTZ cameras [11]
- 3dSOBS+: the enhanced 3D neural model [13]
- RGBD_SOBS: exploiting depth from RGBD videos [22][25]. See here results on the SBM-RGBD dataset dataset

Stopped object detection

- Stopped object detection by neural dual background modeling on GPUs [5]
- Model-based segmentation of stopped foreground objects (SFS) [9]

People counting

- People counting by learning their appearance in a multi-view camera environment [12]

License Plate Recognition

- Video-Based Access Control by Automatic License Plate Recognition [16]

Scene background initialization

- Review [14]
- Taxonomy [19]
- Benchmarking [15], [17], [18], [21]
- Approaches [18], [20]

Main collaborations

The research is mainly conducted in collaboration with the Computer Vision and Pattern Recognition Laboratory (CVPRLab) of the University of Naples Parthenope, headed by [26]. Further fruitful collaborations include

, Université de La Rochelle, France
, University of Bristol, UK
, Université de Sherbrooke, Canada
, Universitat de les Illes Balears, Spain
, Universidad Politécnica de Madrid & Universidad Autónoma de Madrid, Spain

Useful/downloadable material

Software:

- Moving Object Detection Software for RGBD data (RGBD-SOBS): Page created in order to distribute a prototype software implementing the algorithm for Self-organizing background subtraction using color and depth data (RGBD-SOBS) presented in [25].
- Moving Object Detection Software (3dSOBS+): Page created in order to distribute a prototype software implementing the Enhanced 3D Self-Organizing Background Subtraction (3dSOBS+) algorithm presented in [13].
- Moving Object Detection Software (SC-SOBS): Page created in order to distribute a prototype software implementing the Spatially Coherent Self-Organizing Background Subtraction (SC-SOBS) algorithm presented in [8]. New: release with bug fix.
- Moving Object Detection Software (SOBS): Page created in order to distribute a prototype software implementing the Self-Organizing Background Subtraction (SOBS) algorithm presented in [2].

Datasets and results

- SBM-RGBD dataset: a new dataset for moving object detection on RGBD videos, presented in [23], available since 2017.

- SBMnet dataset: a new dataset for Scene Background Modeling, presented in [21], available since 2016.
- Scene Background Initialization (SBI) dataset: Dataset assembled in order to evaluate and compare the results of background initialization algorithms, presented in [15], [17], available since 2015.
- Ground Truth for Sequence Backyard: This .zip file contains the hand-segmented ground truth masks of the sequence Backyard (publicly available at http://www.micc.unifi.it/vim/datasets/ptz-camerapose-recovery/), created for computing some of the PTZ-SOBS performance results presented in [11].
- Stopped Object Detection Sequences: Page created in order to make available image sequences used for testing the Stopped Foreground Subtraction (SFS) algorithm and the 3D Self-Organizing Background Subtraction (3D_SOBS) algorithm presented in [9].
- Moving Object Detection Software (SOBS_CF): Page created in order to distribute results on the www.changedetection.net dataset of a prototype software implementing the Fuzzy Coherence-based Self-Organizing Background Subtraction (SOBS_CF) algorithm presented in [4].
- Moving Object Detection Sequences: Page created in order to show the images used for testing the Self-Organizing Background Subtraction (SOBS) algorithm presented in [2].

Publications on Moving and Stopped Object Detection

[26] L. Maddalena, M. Gori, and S.K. Pal, Pattern recognition and beyond: Alfredo Petrosino's scientific results, Pattern Recognition Letters, Elsevier, DOI: https://doi.org/10.1016/j.patrec.2020.07.032, Vol. 138, 659-669, 2020.

[25] L. Maddalena and A. Petrosino, Self-Organizing Background Subtraction Using Color and Depth Data, Multimedia Tools and Applications 78(9), 11927--11948, Springer, 2019.

[24] L. Maddalena and A. Petrosino, Background Subtraction for Moving Object Detection in RGBD Data: A Survey, J. Imaging 4(5), 71, 2018.

[23] M. Camplani, L. Maddalena, G. Moyá Alcover, A. Petrosino, and L. Salgado, A Benchmarking Framework for Background Subtraction in RGBD Videos, in Battiato S., Farinella G., Leo M., Gallo G. (eds) New Trends in Image Analysis and Processing – ICIAP 2017. Lecture Notes in Computer Science, vol 10590, Springer, DOI: 10.1007/978-3-319-70742-6_21, pp. 219-229, 2017.

[22] L. Maddalena and A. Petrosino, Exploiting Color and Depth for Background Subtraction, in Battiato S., Farinella G., Leo M., Gallo G. (eds) New Trends in Image Analysis and Processing – ICIAP 2017. Lecture Notes in Computer Science, vol 10590, Springer, DOI: 10.1007/978-3-319-70742-6_24, pp. 254--265, 2017.

[21] P. M. Jodoin, L. Maddalena, A. Petrosino and Y. Wang, Extensive Benchmark and Survey of Modeling Methods for Scene Background Initialization, in IEEE Transactions on Image Processing, DOI: 10.1109/TIP.2017.2728181, vol. 26, no. 11, pp. 5244-5256, Nov. 2017.

[20] A. Petrosino, L. Maddalena, T. Bouwmans, Editorial–Scene background modeling and initialization, Pattern Recognition Letters, Elsevier, DOI 10.1016/j.patrec.2017.05.032, Vol. 96, pp. 1-2, 2017.

[19] T. Bouwmans, L. Maddalena, A. Petrosino, Scene background initialization: A taxonomy, Pattern Recognition Letters, Elsevier, DOI 10.1016/j.patrec.2016.12.024, Vol. 96, pp. 3-11, 2017. Free access copy here (available till October 20th, 2017).

[18] L. Maddalena, A. Petrosino, Extracting a Background Image by a Multi-modal Scene Background Model, 23rd International Conference on Pattern Recognition (ICPR), Cancun, DOI: 10.1109/ICPR.2016.7899623, pp. 143-148, 2016.

[17] L. Maddalena, A. Petrosino, Towards Benchmarking Scene Background Initialization, in V. Murino et al. (eds), New Trends in Image Analysis and Processing -- ICIAP 2015 Workshops, Lecture Notes in Computer Science, Vol. 9281, Springer International Publishing Switzerland, DOI 10.1007/978-3-319-23222-5_57#, pp. 469–476, 2015.

[16] E. Di Nardo, L. Maddalena, A. Petrosino, Video-Based Access Control by Automatic License Plate Recognition, in S. Bassis, A. Esposito, and F.C. Morabito (Eds), Advances in Neural Networks: Computational and Theoretical Issues 37, Smart Innovation, Systems and Technologies, Springer International Publishing, 103-117, 2015

[15] L. Maddalena, A. Petrosino, Towards Benchmarking Scene Background Initialization, arXiv:1506.04051, 2015.

[14] L. Maddalena, A. Petrosino, Background Model Initialization for Static Cameras, in T. Bouwmans, F. Porikli, B. Höferlin, and A. Vacavant (Eds), Background Modeling and Foreground Detection for Video Surveillance, DOI: 10.1201/b17223-5, pp. 3-1–-3-16, Chapman and Hall/CRC 2014.

[13] L. Maddalena, A. Petrosino, The 3dSOBS+ algorithm for moving object detection, Computer Vision and Image Understanding, DOI: 10.1016/j.cviu.2013.11.006, vol. 122, pp. 65–73, 125-134, 2014.

[12] L. Maddalena, A. Petrosino, F. Russo, People counting by learning their appearance in a multi-view camera environment, Pattern Recognition Letters, Elsevier, DOI: 10.1016/j.patrec.2013.10.006, vol. 36, pp. 125-134, 2014.

[11] A. Ferone, L. Maddalena, Neural Background Subtraction for Pan-Tilt-Zoom Cameras, IEEE Transactions on Systems, Man, and Cybernetics: Systems, DOI: 10.1109/TSMC.2013.2280121, vol. 44, no. 5, pp. 571-579, 2014.

[10] A. Petrosino, L. Maddalena, P. Pala, et al. (Eds): New Trends in Image Analysis and Processing - ICIAP 2013 Workshops, Naples, Italy, September 2013, Proceedings, Series: Lecture Notes in Computer Science, vol. 8158, Springer, ISBN 978-3-642-41189-2, 2013.

[9] L. Maddalena, A. Petrosino, Stopped Object Detection by Learning Foreground Model in Videos, IEEE Transactions on Neural Networks and Learning Systems, DOI: 10.1109/TNNLS.2013.2242092, vol.24, no.5, pp.723-735, May 2013. See here for additional downloadable material.

[8] L. Maddalena, A. Petrosino, The SOBS algorithm: What are the limits?, 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), DOI: 10.1109/CVPRW.2012.6238922, pp.21-26, 16-21 June 2012.

[7] S.K. Pal, A. Petrosino, L. Maddalena (Eds), Handbook on Soft Computing for Video Surveillance, Chapman & Hall/CRC, ISBN: 9781439856840, 2012.

[6] L. Maddalena, A. Petrosino, Neural Networks in Video Surveillance: A Perspective View, in S.K. Pal, A. Petrosino, L. Maddalena (Eds), Handbook on Soft Computing for Video SurveillanceChapman & Hall/CRC, ISBN: 9781439856840, pp. 59-78, 2012.

[5] G. Gemignani, L. Maddalena, A. Petrosino, Real-time Stopped Object Detection by Neural Dual Background Modeling, in Workshop Proceedings of Euro-Par 2010, Lecture Notes in Computer Science, Springer Berlin/Heidelberg, DOI: 10.1007/978-3-642-21878-1_44, Vol. 6586, pp. 327-334, 2011.

[4] L. Maddalena, A. Petrosino, A Fuzzy Spatial Coherence-based Approach to Background/ Foreground Separation for Moving Object Detection, Neural Computing and Applications, Springer London, DOI 10.1007/s00521-009-0285-8 (Published online on 23 June 2009), Vol. 19, pp. 179–186, 2010. [Impact Factor: 0.627]

[3] L. Maddalena, A. Petrosino, A. Ferone, Object Motion Detection and Tracking by an Artificial Intelligence Approach, International Journal of Pattern Recognition and Artificial Intelligence, Vol. 22, No. 5, World Scientific Publishing Company, Singapore, DOI 10.1142/S0218001408006612, pp. 915-928, 2008. [Impact Factor: 0.66]

[2] L. Maddalena, A. Petrosino, A Self-Organizing Approach to Background Subtraction for Visual Surveillance Applications, IEEE Transactions on Image Processing, DOI 10.1109/TIP.2008.924285, Vol. 17, no. 7, pp. 1168-1177, July 2008. [Impact Factor: 3.315]

[1] L. Maddalena, A. Petrosino, Moving Object Detection for Real-Time Applications, in Proceedings of 14th International Conference on Image Analysis and Processing (ICIAP’07), IEEE Computer Society, Washington, DC, USA, ISBN 0-7695-2877-5, DOI 10.1109/ICIAP.2007.89, pp. 542-547, 2007.


Last update: October 15th, 2020