Background subtraction techniques: a review

By Massimo Piccardi. IEEE International Conference on Systems, Man and Cybernetics, 2004

2,564 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Ocean Engineering

View Paper: https://www.semanticscholar.org/paper/7e4ee9c60bcb1896918ed7debef87f88429855f8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiaogang Wang
2. Trevor Darrell
3. Bernt Schiele
4. Thierry Bouwmans
5. Luc Van Gool
6. Li Fei-Fei
7. Wanli Ouyang
8. Dacheng Tao

In Computer Vision and Pattern Recognition

1. Xiaogang Wang
2. Luc Van Gool
3. Dacheng Tao
4. Yi Yang
5. Ming-Hsuan Yang
6. Bernt Schiele
7. Chunhua Shen
8. Trevor Darrell

In Ocean Engineering

1. Xiaogang Wang
2. Mark Hasegawa-Johnson
3. Niki Trigoni
4. Ioannis Pitas
5. Richard Baraniuk
6. Jenq-Neng Hwang
7. Dikpal Reddy
8. Antonio Brunetti

Citations in the corpus

• Visual SLAM and Structure from Motion in Dynamic Environments (2018, methodology): “Instead of initializing the foreground object, background initialization sets a background model similarly found in background subtraction techniques [7, 126].”
• Intelligent multi-camera video surveillance: A review (2013, positive): “Background subtraction (Piccardi, 2004) is widely used for detecting moving objects in video surveillance with static cameras.”
• Fuzzy human motion analysis: A review (2014, methodology): “The studies on the background subtraction [56, 57] have show n that the Gaussian Mixture Model (GMM) is one of the popular approaches used in modeling the dynamic backg round scene.”
• Traditional and recent approaches in background modeling for foreground detection: An overview (2014, background): “In 2004, Piccardi [201] provided a review on 7 methods and an original categorization based on speed, memory requirements and accuracy.”
• Evaluation of background subtraction techniques for video surveillance (2011, background): “Over the past years, various BS methods have been developed (cf. surveys [24, 9, 4, 2, 3]), with each of them having its own characteristic, strength and weakness.”
• Exploring Weak Stabilization for Motion Feature Extraction (2013, background): “In this case, the observed image motion contains camera-, object-, and part-centric motion.”
• View-Invariant Action Recognition Based on Artificial Neural Networks (2012, methodology): “Moving object segmentation techniques [25], [26] are applied to each action video frame to create binary images depicting person’s body in white and the background in black.”
• Computer vision and deep learning techniques for pedestrian detection and tracking: A survey (2018, methodology): “The initial approach for detection and tracking of moving objects into a video flow acquired by a static camera consisted in the Background Subtraction (BS); this technique allows the detection and distinction of moving objects inside a scene using an appropriate background model [48] .”
• High-throughput Ethomics in Large Groups of Drosophila (2009, background): “Classifying a pixel location as foreground (belonging to a fly) or background (not belonging to any fly) is referred to as background subtraction [5].”
• ViBE: A powerful random technique to estimate the background in video sequences (2009, background): “Index Terms— Surveillance, Pattern recognition, Signal analysis, Video signal processing”
• Unsupervised Detection and Tracking of Arbitrary Objects with Dependent Dirichlet Process Mixtures (2012, methodology): “Often, background subtraction refers to methods that compare an image containing targets with an image of the background only or with some model of the background that is learned as the video progresses [51], while frame differencing refers to methods that compare pairs of consecutive images in a vi”
• People reidentification in surveillance and forensics (2013, background): “Some of these topics include moving target detection, which has been dealt with in surveillance using background subtraction techniques [Piccardi 2004]; people tracking, which exploits time coherency to follow the same object/person along time and space [Yilmaz et al. 2006]; and people detection by…”
• Moving Object Detection by Detecting Contiguous Outliers in the Low-Rank Representation (2011, methodology): “Alternatively , background subtraction [7] compares images with a background model and detects the changes as obj ects.”
• Compressive sensing: From theory to applications, a survey (2013, methodology): “Data captured by single pixel camera can also be used for background subtraction for automatic detection and tracking of objects [57], [58].”
• CDnet 2014: An Expanded Change Detection Benchmark Dataset (2014, background): “Piccardi, 2004 [20] Review of 7 background subtraction methods which highlights their strengths and weaknesses.”
• Advances in View-Invariant Human Motion Analysis: A Review (2010, background): “Generally the reference frame often call ed the “background image” or “background model” must be a representation of the scene without moving objects and must be kept regularly updated so as to adapt to the varying luminance conditions and geometry settings [30].”