Contour tracking by stochastic propagation of conditional density

Isard, Michael; Blake, A.

doi:10.1007/bfb0015549

Cited by 699 publications

(714 citation statements)

References 15 publications

Supporting

Mentioning

686

Contrasting

Unclassified

Order By: Relevance

“…This process is repeated as frames arrive for processing at successive time increments. Unlike [7] we will use such techniques in a situation where f X is not directly proportional to the probability that the image data was created using state X. This means we cannot formally prove that our scheme is asymptotically correct [2], but instead we verify experimentally that our fitness function gives correct results.…”

Section: Introductionmentioning

confidence: 93%

See 1 more Smart Citation

Simultaneous Pose Estimation of Multiple People using Multiple-View Cues with Hierarchical Sampling

Mitchelson¹,

Hilton²

2003

Procedings of the British Machine Vision Conference 2003

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 93%

“…A standard method for doing this is the particle filter [7,2]. This technique maintains a collection of N likely states¨X i : i © 1 N , known as particles.…”

Section: Introductionmentioning

confidence: 99%

Simultaneous Pose Estimation of Multiple People using Multiple-View Cues with Hierarchical Sampling

Mitchelson¹,

Hilton²

2003

Procedings of the British Machine Vision Conference 2003

View full text Add to dashboard Cite

“…However, these equations are intractable for all but certain simple distributions and so approximation methods have to be used. Monte Carlo methods such as particle filtering [5,7] represent one way of evaluating the terms. Our suggested approach, which has shown to be effective, is based on hierarchical partitioning of the state space.…”

Section: Tree-based Filteringmentioning

confidence: 99%

“…Currently, most trackers use particle filtering [7], in which it is essential to be able to sample from the prior to generate new hypotheses. With the increase in computational power one may also consider handling ambiguous situations by treating tracking as object detection in each frame.…”

Section: Introductionmentioning

confidence: 99%

Learning a Kinematic Prior for Tree-Based Filtering

Thayananthan¹,

Stenger²,

Torr³

et al. 2003

Procedings of the British Machine Vision Conference 2003

View full text Add to dashboard Cite

The aim in this paper is to track articulated hand motion from monocular video. Bayesian filtering is implemented by using a tree-based representation of the posterior distribution. Each tree node corresponds to a partition of the state space with piecewise constant density. In a hierarchical search regions with low probability mass can be rapidly discarded, while the modes of the posterior can be approximated to high precision. Large sets of training data are captured using a data glove, and two techniques for constructing the tree are described: One method is to cluster the collected data points using a hierarchical clustering algorithm, and use the cluster centres as nodes. Alternatively, a lower dimensional eigenspace can be partitioned using a grid at multiple resolutions, and each partition centre corresponds to a node in the tree. The effectiveness of these techniques is demonstrated by using them for tracking 3D articulated hand motion in front of a cluttered background.

show abstract

“…Starting from the pioneering formulation of the snake model [8] several attempts to address tracking through the deformation of contours can be found in the literature either model-free [9] or model-based [10]. Level set methods [11] is an established technique [12] to track moving interfaces through model-free [13] or model-based [14] methods with the advantage of being implicit, intrinsic and parameter-free.…”

Section: Introductionmentioning

confidence: 99%

Implicit Free-Form-Deformations for Multi-frame Segmentation and Tracking

Καράντζαλος

Paragios

2005

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper, we propose a novel technique to address motion estimation and tracking. Such technique represents the motion field using a regular grid of thin-plate splines, and the moving objects using an implicit function on the image plane that is a cubic interpolation of a "level set function" defined on this grid. Optical flow is determined through the deformation of the grid and consequently of the underlying image structures towards satisfying the constant brightness constraint. Tracking is performed in similar fashion through the consistent recovery in the temporal domain of the zero iso-surfaces of a level set that is the projection of the Free Form Deformation (FFD) implicit function according to the cubic spline formulation. Such an approach is a compromise between dense motion estimation and parametric motion models, introduces smoothness in an implicit fashion, is intrinsic, and can cope with important object deformations. Promising results demonstrate the potentials of our approach.

show abstract

Contour tracking by stochastic propagation of conditional density

Cited by 699 publications

References 15 publications

Simultaneous Pose Estimation of Multiple People using Multiple-View Cues with Hierarchical Sampling

Simultaneous Pose Estimation of Multiple People using Multiple-View Cues with Hierarchical Sampling

Learning a Kinematic Prior for Tree-Based Filtering

Implicit Free-Form-Deformations for Multi-frame Segmentation and Tracking

Contact Info

Product

Resources

About