BraMBLe: a Bayesian multiple-blob tracker

By Michael Isard, John MacCormick. Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001, 2001

739 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/b46da16dca784e66f600cfa05aa3d9d8bc1dee6d

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Globally-optimal greedy algorithms for tracking a variable number of objects (2011, background): “Classic formulations of multi-object tracking focus on the data association problem of matching instance labels with temporal observations [11, 6, 7, 13 ].”
Maintaining multimodality through mixture tracking (2003, methodology): “The remainder of this section shows how this non-parametric mixture representation can be updated recursively in the same fashion as the two step approach for standard Bayesian sequential estimation.”
Unsupervised Bayesian Detection of Independent Motion in Crowds (2006, methodology): “Isard and MacCormick’s BraMBLe system [6] uses a very simple motion model to help with detection in ambiguous situations.”
Data fusion for visual tracking with particles (2004, methodology): “The model can then be defined in parametric form, using, for instance, mixtures if Gaussians (e.g., [23], [55], [59]).”
Appearance models for occlusion handling (2006, background): “By comparing the difference between the background image and the current image, a given pixel is classified into one of four categories: original background, shaded background or shadow, highlighted background, and foreground objects. thresholds are calculated automatically, details can be found in…”
Segmentation and Tracking of Multiple Humans in Crowded Environments (2008, methodology): “When objects are highly interoccluded, their image observations are far from independent; hence, a joint likelihood for multiple objects is necessary [46], [27], [19], [35], [30], [51].”
Variational inference for visual tracking (2003, background): “Depending on the application a wide variety of objects may be of interest, including people [6], cars [7], etc.”
Color-Based Probabilistic Tracking (2002, methodology): “In the case where a background model is used for multiple object tracking, our Monte Carlo tracker bears connections with the “Bramble” tracker introduced in [11].”
Tracking multiple objects with particle filtering (2002, background): “In the same context, a Bayesian multiple-blob tracker called BraMBLe [14] has just been proposed.”
An MCMC-Based Particle Filter for Tracking Multiple Interacting Targets (2004, background): “BraMBLe [19] addresses tracking and initializing multiple targets in a variable-dimension framework.”
Detection and Tracking of Multiple, Partially Occluded Humans by Bayesian Combination of Edgelet based Part Detectors (2007, neutral): “Because the joint hypotheses space is usually of high dimension, an efficient optimization algorithm, such as a particle filter ( Isard and MacCormick, 2001 ), MCMC (Zhao and Nevatia, 2004a; Smith et al., 2005) or EM (Peter et al., 2005) is used.”
MCMC-based particle filtering for tracking a variable number of interacting targets (2005, methodology): “To model this, we can introduce a new variable, namely, the set of identifiers k t of targets currently in view [18].”
A Boosted Particle Filter: Multitarget Detection and Tracking (2004, background): “1 Introduction Automated tracking of multiple objects is still an open prob lem in many settings, including car surveillance [10], sports [12, 13] and smart rooms [6] am ong many others [5, 7, 11].”
Dissimilarity-based people re-identification and search for intelligent video surveillance (2013, background): “• on-line tracking of the movements of a person or an object of interest [70, 100, 137];”
Counting Crowded Moving Objects (2006, neutral): “For example, we do not plan to solve tracking through occlusions [7], which frees us from complex manual inputs such as initializations [3, 17] or intricate model deﬁnitions [9].”
a CAPpella: programming by demonstration of context-aware applications (2004, methodology): “To determine the number of people in a scene and their location, we built a simplified implementation of a multihypothesis framework – Bramble [13].”
Robust Object Tracking with Online Multiple Instance Learning (2011, neutral): “Although there has been some success with building trackers for specific object classes (e.g., faces [1], humans [2], mice [3], rigid objects [4]), tracking generic objects has remained challenging because an object can drastically change appearance when deforming, rotating out of plane, or when…”