A Probabilistic Exclusion Principle for Tracking Multiple Objects

By John MacCormick, Andrew Blake. Proceedings of the Seventh IEEE International Conference on Computer Vision, 2000

692 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/a2c9926a4bf3f93790dab8bf99c154225b3a1f4a

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

Citations in the corpus

Probabilistic Data Association Methods for Tracking Complex Visual Objects (2001, result): “The JLF is similar to work described in [ 34 ]; a data association approach to tracking is also taken in [35].”
People Tracking with Mobile Robots Using Sample-Based Joint Probabilistic Data Association Filters (2003, methodology): “They have been applied with great success to different state estimation problems including visual tracking (Black and Jepson 1998; Isard and Blake 1996), mobile robot localization (Dellaert et al. 1999; Fox et al. 1999, 2001; Gross, Boehme, and König 2002; Lenser and Veloso 2000; Montemerlo et al. 2”
Multi-camera multi-person tracking for EasyLiving (2000, background): “MacCormick and Blake[17] describe a modification of CONDENSATION tracking that incorporates an exclusion principle to keep multiple head tracks from coalescing onto one head.”
Kernel-Based Object Tracking (2003, background): “Probabilistic exclusion for tracking multiple objects was discussed in [51].”
Video Primal Sketch: A Unified Middle-Level Representation for Video (2015, methodology): “In higher level object motion tracking, people represented different tracking units depending on the underlying objects and scales, such as sparse or dense feature points tracking [4,32], kernels tracking [10,16], contours tracking [24], and middlelevel pairwise-components generation [41].”
Structured Determinantal Point Processes (2010, background): “We would like a distribution over sets of trajectories that (1) includes sets of different cardinality and (2) prefers sets of trajectories that are spread out in space-time, as objects are likely to be [11, 15].”
A Novel Stastical Particle Filtering Approach for Non- Linear and Non-Gaussian System Identification (2012, methodology): “This sequential MC (SMC) approach is known variously as bootstrap filtering [3], the condensation algorithm [6], particle filtering [7], interacting particle approximations [8], [9], and survival of the fittest [10].”
Object tracking: A survey (2006, background): “Particle .ltering [MacCormick and Blake 2000] M F v Gradient mag.”
An adaptive color-based particle filter (2003, methodology): “Our work has evolved from the Condensation algorithm [12] [14] which was developed in the computer vision community and was typically used with edge-based image features [11] [12] [14] [20] .”
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].”
SMOOTH VARIABLE STRUCTURE FILTERING: THEORY AND APPLICATIONS (2011, background): “The PF may also be known as Monte Carlo filters [31], interacting particle approximations [74], bootstrap filters [75], condensation algorithms [76], and survival of the fittest methods [77], to name a few.”
Articulated body motion capture by annealed particle filtering (2000, methodology): “Partitioned sampling was developed by MacCormick and Blake [13] as a variation on Condensation to reduce the number of particles needed to track more than one object.”
A Gesture Based Interface for Human-Robot Interaction (2000, background): “Recent progress of head tracking with occlusion MacCormick and Blake, 1999 might provide an answer.”
Tracking multiple moving targets with a mobile robot using particle filters and statistical data association (2001, methodology): “One way to apply particle filters to the problem of tracking multipleobjects is to estimate the combined state space, as proposed in [11].”
Sequential Monte Carlo methods for multiple target tracking and data fusion (2002, methodology): “In computer vision, a probabilistic exclusion principle has been developed in [24] to track multiple objects, but the algorithm is very dependent of the observation model and is only applied for two objects.”
Monocular Pedestrian Detection: Survey and Experiments (2009, background): “Extensions that are especially relevant for pedestrian tracking involve hybrid discrete/continuous state-spaces [26], [50] and efficient sampling strategies [13], [32], [36], [44].”
Fast multiple object tracking via a hierarchical particle filter (2005, methodology): “The particle filter based tracking algorithms usually use contours, color features, or appearance models [17, 18, 25, 37, 20].”
MCMC-based particle filtering for tracking a variable number of interacting targets (2005, background): “Instead of clustering particles, the probabilistic exclusion principle approach [23] adds a term to the measurement model that assures that every feature measured belongs to only one target.”