Overview
Author: Thierry BOUWMANS, Associate Professor, Lab. MIA, Univ. La Rochelle, France.
A full overview of the background subtraction methods listed in this website are provided in:
Editors: T. Bouwmans, F. Porikli, B. Hörferlin, A. Vacavant.
Title: Handbook “Background modeling and Foreground Detection for video surveillance: Traditional and Recent Approaches, Benchmarking and Evaluation".
Publisher :CRC Press, Taylor and Francis Group.
Publication Date : July 1, 2014. (More information) [Purchase]
Further Improvements
If you would like to list your publication related to this topic on this website, please send me your publication in .pdf and I will add the reference.
Fair Use Policy
As this website gives many information that come from my research, please cite my following survey papers:
T. Bouwmans, “Traditional and Recent Approaches in Background Modeling for Foreground Detection: An Overview”, Computer Science Review, 2014. [pdf]
T. Bouwmans, E. Zahzah, “Robust PCA via Principal Component Pursuit: A Review for a Comparative Evaluation in Video Surveillance”, Special Issue on Background Models Challenge, Computer Vision and Image Understanding, CVIU 2014, Volume 122, pages 22–34, May 2014. [pdf]
T. Bouwmans, "Recent Advanced Statistical Background Modeling for Foreground Detection: A Systematic Survey", Recent Patents on Computer Science, Volume 4, No. 3, pages 147-176, September 2011. [pdf]
My recent publications are available on Academia, Research Gate, Researchr, ORCID and Publication List.
Objective
The aim of this web site is to provide resources such as references (3734 papers), datasets (36 datasets), codes (80 codes) and links to demonstration websites (65 websites) for the research on background subtraction by grouping all related researches and particularly recent advances in this field. For this, it is organized in the following sections:
1. Background Modeling (2336 papers): This section groups all the references that I found on background modeling. These references are listed in the following categories:
Traditional Background Modeling (861 papers)
Basic Background Modeling (43 papers), Statistical Background Modeling (603 papers),
Background Modeling via Clustering (109 papers),
Neural Network Background Modeling (81 papers), Background Estimation (25 papers).
Recent Background Modeling (1438 papers)
A) Advanced Miscellaneous Models [Unsupervised] (197 papers)
Sample based Background Modeling (53 papers) , Advanced Statistical Background Modeling (100 papers),
Fuzzy Background Modeling (40 papers)
Background Modeling via Circulant Structures (2 papers)
B) Deep Learning Models [Supervised]
Background Modeling via Deep Learning (209 papers)
C) Zero Shot Learning Models [Unsupervised]
Background Modeling via Zero Shot Learning (1 paper)
D) Subspace Learning/Clustering Models [Unsupervised] (989 papers)
Background Modeling via Discriminative Subspace Learning (3 papers),
Background Modeling via Mixed Subspace Learning (1 paper),
Background Modeling via Kernel Subspace Learning (1 paper),
Background Modeling via Subspace Tracking (21 papers),
Background Modeling via RSL (5 papers),
Background via other Robust Subspace Models (10 papers),
Background Modeling via RLR (1 paper), Background Modeling via RLMC (1 paper), Background Modeling via RST (9 papers)
Background Modeling by RPCA (538 papers), Background Modeling via RNMF (6 papers),
Background Modeling via RMC (27 papers),
Background Modeling via other Robust Subspace Recovery (9 papers)
Background Modeling via RLRM (59 papers), Background Modeling via RGL (1 paper),
Background Modeling via Robust Graphical Lasso (1 paper)
Background Modeling via Subspace Tracking (21 papers),
Background Modeling via Multiple Subspace Tracking (4 papers)
Background Modeling via RTSL (1 paper), Background Modeling via CRTC (6 papers),
Background Modeling TLR (1 paper),
Background Modeling via RPCA Tensors (136 papers), Background Modeling via RNMF Tensors (2 papers)
Background Modeling via RTC (19 papers), Background Modeling via other Robust Tensor Recovery (6 papers),
Background Modeling via LRT (10 papers), Background Modeling via Tensor Subspace Tracking (5 papers),
Background Modeling with Superposed Atomic Representation (1)
Sparse Background Modeling (83 papers),
Background Modeling via Subspace Clustering (1 paper)
E) Signal Processing Models (42 papers)
Transform Domain Background Modeling (37 papers), Graph Signal Processing Modeling (5 papers)
Prospective Background Modeling (37 papers)
Pixel Intensity Classification (PIC) (5 papers), Pixel Change Classification (PCC) (1 paper)
Co-occurrence (CO) (9 papers), Pursuing Dynamic Spatio-Temporals Models (STDM) (3 paper),
Projection Pursuit Density Estimation (PPD) (1 paper), Sample-based Models (2 papers),
Confidence Measure Background Modeling (3 papers), Linear Model (2 papers),
Linear Dynamical System (1 paper), Data-Driven Control (3 papers)
Nested Models (1 paper), Polynomial Models (4 papers), PID Models (1 paper),
Imbalanced Learning Models (1 paper)
2. Background Initialization (105 papers): This section groups all the references that I found on background initialization. These references are listed in the following categories:
Temporal Statistics (TS) (21 papers), Subsequences of Stable Intensity (SSI) (21 papers),
Iterative Model Completion (IMC) (19 papers),
Optimal Labeling (OL) (2 papers), Missing Data Reconstruction (MDR) (15 papers),
Neural Networks (NN) (25 papers), Image Fusion (IF) (1 paper), Particle Shape (1)
3. Background Maintenance (15 papers): This section groups all the references that I found on background maintenance. These references are listed in the following categories:
Maintenance rules (2 papers), Learning rates (10 papers), Adaptation rates (3 papers)
4. Foreground Detection (44 papers)
4.1 Foreground Detection (29 papers): This section groups all the references that I found on foreground detection. These references are listed in the following categories:
Basic Foreground Detection, Statistical Foreground Detection (4 papers), Fuzzy Foreground Detection (25 papers), Foreground Detection - Others
4.2 . Post-Processing (15 papers)
Basic Post-processing (5 papers), Statistical Post-processing (1 paper), Fuzzy Post-processing (5 papers), Deep Learned Post-processing (3 papers), Advanced Post-processing (1 paper)
5. Features (534 papers):This section groups all the references that I found on the different features used for background modeling.
5.1 Feature Size (45 papers)
Pixel, Keypoint (2 papers), Superpixel (13 papers), Block (2 papers), Cluster (4 papers), Region (9 papers), Brick (5 papers), Patch (4 papers), Semantic (6 papers)
5.2 Feature Type (410 papers)
5.2.1 Features in the Pixel Domain (360 papers)
a) Crisp Features (302 papers)
Intensity Features (39 papers), Color Features in Well Known Color Spaces (18 papers),
Color Features in Designed Shape Color Spaces (13 papers)
Designed Illumination Invariant Color Features (6 papers), Color Filter Array (CFA) Features (1 paper), Statistics on Color Features (10 papers),
Edge Features (23 papers), Textures Features (145 papers), Motion Features (2 papers), Stereo Features (45 papers)
b) Statistical Features (26 papers)
Histogram of Intensity (1 paper), Histograms of color space (9 papers), Histograms of Gradient (7 papers),
Histograms of Figure/Ground Segmentations (HFG) (1 paper) ,
Entropy (4 papers), Statistical Reach Feature (2 papers), Statistical Texture (2 papers)
c) Fuzzy Features (16 papers)
Local Fuzzy Color Histograms (LFCH) (6 papers),
Local Fuzzy Color Difference Histograms (LFCDH) (1 paper), Fuzzy Histon (FH) (3 paper),
Fuzzy Correlogram (FC) (2 papers), Fuzzy Statistical Texture (FST) (3 papers),
Fuzzy Local Binary Pattern (FLBP) (1 paper)
d) Other Features (16 papers)
Frequency Features (2 papers), Phase features (5 papers), Spatial features (1 paper), Location features (7 papers), Bit Plane features (1 paper)
5.2.2 Features in a Transform Domain (32 papers)
Frequency Features (21 papers), Video Features (11 papers)
5.2.3. Features in Deep Learning Domain (17 papers)
Deep Learned Features (17 papers)
5.3 Feature Strategies (65 papers)
5.4 Feature Selection (9 papers)
5.5 Feature Reliability (4 papers)
5.6 Feature Relevance (2 papers)
6. Strategies (152 papers): This section groups all the references that I found on the different strategies used in background modeling.
6.1 Multiple Strategies (125 papers)
Multi-Features (65 papers), Multi-Scales (7 papers), Multi-Levels (11 papers), Multi-Resolutions (4 papers),
Multi-Backgrounds (19 papers), Multi-Layers (10 papers), Multi-Classifiers (3 papers), Multi-Cues (5 papers),
Multi-Criteria (1 paper)
6.2 Map Strategies (2 papers)
Map Selection (2 papers)
6.3 Hierarchical Strategies (9 papers)
Hierarchical Strategies (9 papers)
6.4 Coarse to Fine Strategies (3 papers)
Coarse to Fine Strategies (3 papers)
6.5 Compositional Strategies (1 paper)
Compositional Strategies (1 paper)
6.6 Consensus Strategies (3 papers)
Consensus Strategies (3 papers)
6.7 Markov Random Fields (9 papers)
Markov Random Fields (9 papers)
7. Performance Evaluation (23 papers): This section groups all the references that I found on the different method to evaluate background modeling methods.
Background subtraction (21 papers), Background initialization (2 papers)
Additional Informations
In addition of the previous sections, I give references and links to:
Surveys (23 papers)
Datasets Generation (1 paper)
Available Datasets (41 datasets)
Intelligent Visual Surveillance of Human Activities (34 datasets), Intelligent Visual Observation of Animals and Insects (7 datasets)
Results on the Wallflower Dataset for Statistical Background Modeling
First Generation , Second Generation, Third Generation
Available implementations (89 implementations)
1) Traditional Background Modeling (19 implementations)
Basic Background Modeling (2), Statistical Background Modeling (12), Background Modeling via Clustering (2),
Neural Network Background Modeling (3), Background Estimation (0)
2) Recent Background Modeling (59 implementations)
Advanced Background Modeling (3), Advanced Statistical Background Modeling (2),
Deep Learning Bakground Modeling (9), Fuzzy Background Modeling (1),
Background Modeling via Subspace Tracking (1), Background Modeling via RSL (1),
Background Modeling by Robust PCA (23), Background Modeling via RNMF (2),
Background Modeling via RMC (2), Background Modeling via other Robust Subspace Recovery (1)
Background Modeling via Subspace Tracking (5) , Background Modeling via LRM (4),
Background Modeling via RPCA Tensors (1), Sparse Background Modeling (2),
Transform Domain Background Modeling (0), Graph Signal Processing Background Modeling (1)
3) Others Background Modeling (3 implementations)
Others Background Modeling (3 implementations)
4) Background Initialization (5 implementations)
Background Initialization (5 implementations)
5) Background Maintenance
6) Foreground Detection (3 implementations)
Foreground Detection (3 implementations)
7) Background Modeling - Codes : Background Modeling - Packages (3 packages) - Background Modeling - Librairies (2 librairies)
Real Time Implementations (144 papers)
GPU Implementations (42 papers), Embedded Implementations (38 papers), Architecture Implementations (53 papers), Parallel implementations (9 papers), Quantum Computing (1 paper), Near-Sensor Implementations (1 paper)
Sensors (111 papers)
Cameras (23 papers), Infrared Cameras (32 papers), Multispectral Cameras (2 papers), RGB-D Cameras (41 papers), Light Field Cameras (4 papers), Radar Systems (9 papers)
Journals - Conferences - Workshops
This page lists Journals (7), Conferences (16), Workshops (5) where recent advances on background subtraction are mainly published.
Websites (66 Websites)
1) Traditional Background Modeling (13 websites)
Basic Background Modeling (0), Statistical Background Modeling (9), Background Modeling via Clustering (1),
Neural Network Background Modeling (2), Background Estimation (1)
2) Recent Background Modeling (38 websites)
Advanced Statistical Background Modeling (2), Fuzzy Background Modeling (1),
Background Modeling via Discriminative Subspace Learning (1) ,
Background Modeling via Mixed Subspace Learning (0)
Background Modeling via Kernel Subspace Learning (0), Background Modeling via RSL (1),
Background Modeling by Robust PCA (22), Background Modeling via RNMF (0),
Background Modeling via RMC (1), Background Modeling via other Robust Subspace Recovery (0)
Background Modeling via Subspace Tracking (4), Background Modeling via LRM (4),
Sparse Background Modeling (2), Transform Domain Background Modeling (0).
3) Others Background Modeling (10 websites)
Others Background Modeling (10)
4) Background Initialization: Background Initialization (4)
5) Background Maintenance
6) Foreground Detection: Foreground Detection (3)
7) Performance Evaluation: Performance Evaluation (1)
Applications (184 papers)
Intelligent Video Surveillance (68 papers), Intelligent Visual Observation of Animals and Insects (55 papers),
Intelligent Surveillance of Natural Environments (30 papers), Pose Estimation (1 paper),
Optical Motion Capture (7 papers),
Human Computer Interaction (2 papers), Content based Video Coding (21 papers)
Extension to moving cameras (60 papers)
Freely Moving Cameras (7 papers), Online Moving Cameras (2 papers), Moving Cameras (30 papers),
Hand-held Cameras (1 paper), Mobile Cameras (2 papers), UAV Cameras (1 paper), PTZ Cameras (17 papers)
Patents (1 patent)
Note: The number in parenthesis gives the total amount of papers or links for the related categories.
Success Story
You can see the activity of the Background Subtraction Web Site at this page Google Analytics.