Background modeling and subtraction of dynamic scenes

By Antoine Monnet, Anurag Mittal, Nikos Paragios, Visvanathan Ramesh. Proceedings Ninth IEEE International Conference on Computer Vision, 2003

482 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/af444713a0cbbc12773a3a966b0fdd340f83b88f

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

A texture-based method for modeling the background and detecting moving objects (2006, methodology): “Unlike most other approaches, the features in background modeling are”
Real-time foreground-background segmentation using codebook model (2005, background): “1077-2014/$ - see front matter r 2005 Elsevier Ltd.”
Video Processing From Electro-Optical Sensors for Object Detection and Tracking in a Maritime Environment: A Survey (2016, methodology): “Autoregressive updates of background models is done by identifying the subspace of consistently recurring backgrounds [100], [120], [137], [162].”
Segmenting Salient Objects from Images and Videos (2010, background): “Perhaps the most common approach to reduce scene clutter is to detect moving objects against a static background [8, 9 ,10].”
Animal Detection From Highly Cluttered Natural Scenes Using Spatiotemporal Object Region Proposals and Patch Verification (2016, methodology): “The use of multiple hypotheses to describe the behavior of an evolving scene at the pixel level significantly improves the performance of background modeling and subtraction [15].”
Moving Object Detection in Spatial Domain using Background Removal Techniques - State-of-Art (2008, background): “[75] present an autoregressive form to predict the frame to be observed.”
Moving Object Detection by Detecting Contiguous Outliers in the Low-Rank Representation (2011, methodology): “Typical me thods include single Gaussian distribution [27], Mixture of Gaussian [28], kernel densit y estimation [29], [30], block correlation [31], codebook model [32], Hidden Markov model [33], [34] an d linear autoregressive models [8], [35], [36].”
Performance evaluation of object detection algorithms for video surveillance (2006, methodology): “To cope with abrupt changes, multiple model techniques have been proposed [18] as well as predictive stochastic models (e.g., AR, ARMA [24], [25]).”
A survey on activity recognition and behavior understanding in video surveillance (2013, background): “[111] autoregression i,o g s m handle position change in light source and local/global illumination changes and fail to detect moving object in same direction of wind flow”
Improved adaptive Gaussian mixture model for background subtraction (2004, background): “For example in [5, 11] the pixel value distribution over time is modelled as an autoregressive process.”
Background-subtraction using contour-based fusion of thermal and visible imagery (2007, methodology): “Kalman filters were used in [55], and an auto-regressive model was used in [32].”
Bayesian modeling of dynamic scenes for object detection (2005, methodology): “The local search proposed in [7], the SDG of [29], the time series models of [25], [40], and KDEs over color and optical flow in [24] are several formulations proposed for detection nonstationary backgrounds .”
Object tracking: A survey (2006, background): “1999], Dynamic texture background [Monnet et al. 2003].”
Spatiotemporal Saliency in Dynamic Scenes (2010, background): “Shortcomings of even the most advanced techniques include the requirements of 1. static cameras [12], [21], [25]; 2. explicit [14], or approximate [23], compensation of camera motion; 3. foreground objects that move in a consistent direction (an assumption that we denote as temporal coherence) [6],…”
Motion-based background subtraction using adaptive kernel density estimation (2004, methodology): “One can observe that even recent predictive methods do not model multiple modalities of dynamic behavior [20] [26] [23], and therefore such models can be quite inefficient in many practical scenarios.”
Human activities recognition using depth images (2013, background): “Human activities recognition is useful in many applications like surveillance, action/event centric video retrieval and patient monitoring systems.”
Efficient adaptive density estimation per image pixel for the task of background subtraction (2006, background): “For example, in (Toyama et al., 1999; Monnet et al., 2003) the pixel value distribution over time is modelled as an autoregressive process.”
Background Subtraction for Freely Moving Cameras (2009, background): “[20], Zhong and Sclaroff [38], and Mittal and Paragios [19]).”