Results 1  10
of
22
Contour Tracking By Stochastic Propagation of Conditional Density
, 1996
"... . In Proc. European Conf. Computer Vision, 1996, pp. 343356, Cambridge, UK The problem of tracking curves in dense visual clutter is a challenging one. Trackers based on Kalman filters are of limited use; because they are based on Gaussian densities which are unimodal, they cannot represent s ..."
Abstract

Cited by 565 (23 self)
 Add to MetaCart
. In Proc. European Conf. Computer Vision, 1996, pp. 343356, Cambridge, UK The problem of tracking curves in dense visual clutter is a challenging one. Trackers based on Kalman filters are of limited use; because they are based on Gaussian densities which are unimodal, they cannot represent simultaneous alternative hypotheses. Extensions to the Kalman filter to handle multiple data associations work satisfactorily in the simple case of point targets, but do not extend naturally to continuous curves. A new, stochastic algorithm is proposed here, the Condensation algorithm  Conditional Density Propagation over time. It uses `factored sampling', a method previously applied to interpretation of static images, in which the distribution of possible interpretations is represented by a randomly generated set of representatives. The Condensation algorithm combines factored sampling with learned dynamical models to propagate an entire probability distribution for object pos...
Icondensation: Unifying lowlevel and highlevel tracking in a stochastic framework
, 1998
"... . Tracking research has diverged into two camps; lowlevel approaches which are typically fast and robust but provide little finescale information, and highlevel approaches which track complex deformations in highdimensional spaces but must trade off speed against robustness. Realtime highlevel ..."
Abstract

Cited by 259 (13 self)
 Add to MetaCart
. Tracking research has diverged into two camps; lowlevel approaches which are typically fast and robust but provide little finescale information, and highlevel approaches which track complex deformations in highdimensional spaces but must trade off speed against robustness. Realtime highlevel systems perform poorly in clutter and initialisation for most highlevel systems is either performed manually or by a separate module. This paper presents a new technique to combine low and highlevel information in a consistent probabilistic framework, using the statistical technique of importance sampling combined with the Condensation algorithm. The general framework, which we term Icondensation, is described, and a hand tracker is demonstrated which combines colour blobtracking with a contour model. The resulting tracker is robust to rapid motion, heavy clutter and handcoloured distractors, and reinitialises automatically. The system runs comfortably in real time on an...
Image Parsing: Unifying Segmentation, Detection, and Recognition
, 2005
"... In this paper we present a Bayesian framework for parsing images into their constituent visual patterns. The parsing algorithm optimizes the posterior probability and outputs a scene representation in a "parsing graph", in a spirit similar to parsing sentences in speech and natural language. The ..."
Abstract

Cited by 160 (18 self)
 Add to MetaCart
In this paper we present a Bayesian framework for parsing images into their constituent visual patterns. The parsing algorithm optimizes the posterior probability and outputs a scene representation in a "parsing graph", in a spirit similar to parsing sentences in speech and natural language. The algorithm constructs the parsing graph and reconfigures it dynamically using a set of reversible Markov chain jumps. This computational framework integrates two popular inference approaches  generative (topdown) methods and discriminative (bottomup) methods. The former formulates the posterior probability in terms of generative models for images defined by likelihood functions and priors. The latter computes discriminative probabilities based on a sequence (cascade) of bottomup tests/filters.
Learning to Track the Visual Motion of Contours
 Artificial Intelligence
, 1995
"... A development of a method for tracking visual contours is described. Given an "untrained" tracker, a trainingmotion of an object can be observed over some extended time and stored as an image sequence. The image sequence is used to learn parameters in a stochastic differential equation model. Thes ..."
Abstract

Cited by 97 (16 self)
 Add to MetaCart
A development of a method for tracking visual contours is described. Given an "untrained" tracker, a trainingmotion of an object can be observed over some extended time and stored as an image sequence. The image sequence is used to learn parameters in a stochastic differential equation model. These are used, in turn, to build a tracker whose predictor imitates the motion in the training set. Tests show that the resulting trackers can be markedly tuned to desired curve shapes and classes of motions. Contents 1 Introduction 2 2 Tracking framework 2 2.1 Curve representation : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 2 2.2 Tracking as estimation over time : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 3 2.3 Rigid body transformations : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 3 2.4 Curves in motion : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 7 2.5 Discretetime model : : : : : : : : : :...
Diffeomorphisms Groups and Pattern Matching in Image Analysis
, 1995
"... . In a previous paper, the author proposes to see the deformations of a common pattern as the action of an infinite dimensional group. We show in this paper that this approach can be applied numerically for pattern matching in image analysis of digital images. Using Lie group ideas, we construct a d ..."
Abstract

Cited by 90 (9 self)
 Add to MetaCart
. In a previous paper, the author proposes to see the deformations of a common pattern as the action of an infinite dimensional group. We show in this paper that this approach can be applied numerically for pattern matching in image analysis of digital images. Using Lie group ideas, we construct a distance between deformations defined through a metric given the cost of infinitesimal deformations. Then we propose a numerical scheme to solve a variational problem involving this distance and leading to a suboptimal pattern matching. Contents 1. Introduction 1 2. Algorithmic side: gradient descent on AB 5 3. Numerical scheme 7 4. Numerical results 8 5. Conclusion 15 References 15 1. Introduction In [6, 5], we proposed an infinite dimensional group approach for physics based models in pattern recognition. Let us recall the outlines of this approach in the particular case of image analysis we are interested in. Consider that the set of gray The author would like to thank professor Robert...
Modelling and interpretation of architecture from several images
"... The modelling of 3dimensional (3D) environments has become a requirement for many applications in engineering design, virtual reality, visualisation and entertainment. However the scale and complexity demanded from such models has risen to the point where the acquisition of 3D models can require a ..."
Abstract

Cited by 83 (6 self)
 Add to MetaCart
The modelling of 3dimensional (3D) environments has become a requirement for many applications in engineering design, virtual reality, visualisation and entertainment. However the scale and complexity demanded from such models has risen to the point where the acquisition of 3D models can require a vast amount of specialist time and equipment. Because of this much research has been undertaken in the computer vision community into automating all or part of the process of acquiring a 3D model from a sequence of images. This thesis focuses specifically on the automatic acquisition of architectural models from short image sequences. An architectural model is defined as a set of planes corresponding to walls which contain a variety of labelled primitives such as doors and windows. As well as a label defining its type, each primitive contains parameters defining its shape and texture. The key advantage of this representation is that the model defines not only geometry and texture, but also an interpretation of the scene. This is crucial as it enables reasoning about the scene; for instance, structure and texture can be inferred in areas of the model which are unseen in any
D Position, Attitude and Shape Input Using Video Tracking of Hands and Lips
"... Recent developments in videotracking allow the outlines of moving, natural objects in a videocamera input stream to be tracked live, at full videorate. Previous systems have been available to do this for specially illuminated objects or for naturally illuminated but polyhedral objects. Other syst ..."
Abstract

Cited by 70 (12 self)
 Add to MetaCart
Recent developments in videotracking allow the outlines of moving, natural objects in a videocamera input stream to be tracked live, at full videorate. Previous systems have been available to do this for specially illuminated objects or for naturally illuminated but polyhedral objects. Other systems have been able to track nonpolyhedral objects in motion, in some cases from live video, but following only centroids or keypoints rather than tracking whole curves. The system described here can track accurately the curved silhouettes of moving nonpolyhedral objects at framerate, for example hands, lips, legs, vehicles, fruit, and without any special hardware beyond a desktop workstation and a videocamera and framestore. The new algorithms are a synthesis of methods in deformable models, Bspline curve representation and control theory. This paper shows how such a facility can be used to turn parts of the body  for instance, hands and lips  into input devices. Rigid motion of a...
A smoothing filter for Condensation
 In Proc. European Conf. on Computer Vision
, 1998
"... . Condensation, recently introduced in the computer vision literature, is a particle filtering algorithm which represents a tracked object 's state using an entire probability distribution. Clutter can cause the distribution to split temporarily into multiple peaks, each representing a different ..."
Abstract

Cited by 58 (2 self)
 Add to MetaCart
. Condensation, recently introduced in the computer vision literature, is a particle filtering algorithm which represents a tracked object 's state using an entire probability distribution. Clutter can cause the distribution to split temporarily into multiple peaks, each representing a different hypothesis about the object configuration. When measurements become unambiguous again, all but one peak, corresponding to the true object position, die out. While several peaks persist estimating the object position is problematic. "Smoothing" in this context is the statistical technique of conditioning the state distribution on both past and future measurements once tracking is complete. After smoothing, peaks corresponding to clutter are reduced, since their trajectories eventually die out. The result can be a much improved stateestimate during ambiguous timesteps. This paper implements two algorithms to smooth the output of a Condensation filter. The techniques are derived fro...
SpaceTime Tracking
 IN EUROPEAN CONFERENCE ON COMPUTER VISION
, 2002
"... We propose a new tracking technique that is able to capture nonrigid motion by exploiting a spacetime rank constraint. Most tracking methods use a prior model in order to deal with challenging local features. The model usually has to be trained on carefully handlabeled example data before the trac ..."
Abstract

Cited by 53 (2 self)
 Add to MetaCart
We propose a new tracking technique that is able to capture nonrigid motion by exploiting a spacetime rank constraint. Most tracking methods use a prior model in order to deal with challenging local features. The model usually has to be trained on carefully handlabeled example data before the tracking algorithm can be used. Our new modelfree tracking technique can overcome such limitations. This can be achieved in redefining the problem. Instead of first training a model and then tracking the model parameters, we are able to derive trajectory constraints first, and then estimate the model. This reduces the search space significantly and allows for a better feature disambiguation that would not be possible with traditional trackers. We demonstrate that sampling in the trajectory space, instead of in the space of shape configurations, allows us to track challenging footage without use of prior models.
A Hierarchical Statistical Framework for the Segmentation of Deformable Objects in Image Sequences
, 1994
"... In this paper, we propose a new statistical framework for modeling and extracting 2D moving deformable objects from image sequences. The object representation relies on a hierarchical description of the deformations applied to a template. Global deformations are modeled using a Karhunen Loeve expans ..."
Abstract

Cited by 29 (8 self)
 Add to MetaCart
In this paper, we propose a new statistical framework for modeling and extracting 2D moving deformable objects from image sequences. The object representation relies on a hierarchical description of the deformations applied to a template. Global deformations are modeled using a Karhunen Loeve expansion of the distorsions observed on a representative population. Local deformations are modeled by a (firstorder) Markov process. The optimal bayesian estimate of the global and local deformations is obtained by maximizing a nonlinear joint probability distribution using stochastic and deterministic optimization techniques. The use of global optimization techniques yields robust and reliable segmentations in adverse situations such as low signaltonoise ratio, nongaussian noise or occlusions. Moreover, no human interaction is required to initialize the model. The approach is demonstrated on synthetic as well as on realworld image sequences showing moving hands with partial occlusions.