Color-Based Probabilistic Tracking
28 May 2002-pp 661-675
TL;DR: This work introduces a new Monte Carlo tracking technique based on the same principle of color histogram distance, but within a probabilistic framework, and introduces the following ingredients: multi-part color modeling to capture a rough spatial layout ignored by global histograms, incorporation of a background color model when relevant, and extension to multiple objects.
Abstract: Color-based trackers recently proposed in [3,4,5] have been proved robust and versatile for a modest computational cost They are especially appealing for tracking tasks where the spatial structure of the tracked objects exhibits such a dramatic variability that trackers based on a space-dependent appearance reference would break down very fast Trackers in [3,4,5] rely on the deterministic search of a window whose color content matches a reference histogram color modelRelying on the same principle of color histogram distance, but within a probabilistic framework, we introduce a new Monte Carlo tracking technique The use of a particle filter allows us to better handle color clutter in the background, as well as complete occlusion of the tracked entities over a few framesThis probabilistic approach is very flexible and can be extended in a number of useful ways In particular, we introduce the following ingredients: multi-part color modeling to capture a rough spatial layout ignored by global histograms, incorporation of a background color model when relevant, and extension to multiple objects
Citations
More filters
TL;DR: A new approach toward target representation and localization, the central component in visual tracking of nonrigid objects, is proposed, which employs a metric derived from the Bhattacharyya coefficient as similarity measure, and uses the mean shift procedure to perform the optimization.
Abstract: A new approach toward target representation and localization, the central component in visual tracking of nonrigid objects, is proposed. The feature histogram-based target representations are regularized by spatial masking with an isotropic kernel. The masking induces spatially-smooth similarity functions suitable for gradient-based optimization, hence, the target localization problem can be formulated using the basin of attraction of the local maxima. We employ a metric derived from the Bhattacharyya coefficient as similarity measure, and use the mean shift procedure to perform the optimization. In the presented tracking examples, the new method successfully coped with camera motion, partial occlusions, clutter, and target scale variations. Integration with motion filters and data association techniques is also discussed. We describe only a few of the potential applications: exploitation of background information, Kalman tracking using motion models, and face tracking.
4,996 citations
Cites background from "Color-Based Probabilistic Tracking"
...tracking based on the metric (6) is implemented in [57]....
[...]
23 Jun 2013
TL;DR: Large scale experiments are carried out with various evaluation criteria to identify effective approaches for robust tracking and provide potential future research directions in this field.
Abstract: Object tracking is one of the most important components in numerous applications of computer vision. While much progress has been made in recent years with efforts on sharing code and datasets, it is of great importance to develop a library and benchmark to gauge the state of the art. After briefly reviewing recent advances of online object tracking, we carry out large scale experiments with various evaluation criteria to understand how these algorithms perform. The test image sequences are annotated with different attributes for performance evaluation and analysis. By analyzing quantitative results, we identify effective approaches for robust tracking and provide potential future research directions in this field.
3,828 citations
Cites background or methods from "Color-Based Probabilistic Tracking"
...On the other hand, stochastic search algorithms such as particle filters [28, 44] have been widely used since they are...
[...]
...Some trackers perform better when the scale factor is smaller, such as L1APG, MTT, LOT and CPF....
[...]
...CPF [44] L, IH PF N C 109 LOT [43] L, color PF Y M 0....
[...]
...2Some source codes [36, 58] are obtained from direct contact, and some methods are implemented on our own [44, 16]....
[...]
TL;DR: An extensive evaluation of the state-of-the-art online object-tracking algorithms with various evaluation criteria is carried out to identify effective approaches for robust tracking and provide potential future research directions in this field.
Abstract: Object tracking has been one of the most important and active research areas in the field of computer vision. A large number of tracking algorithms have been proposed in recent years with demonstrated success. However, the set of sequences used for evaluation is often not sufficient or is sometimes biased for certain types of algorithms. Many datasets do not have common ground-truth object positions or extents, and this makes comparisons among the reported quantitative results difficult. In addition, the initial conditions or parameters of the evaluated tracking algorithms are not the same, and thus, the quantitative results reported in literature are incomparable or sometimes contradictory. To address these issues, we carry out an extensive evaluation of the state-of-the-art online object-tracking algorithms with various evaluation criteria to understand how these methods perform within the same framework. In this work, we first construct a large dataset with ground-truth object positions and extents for tracking and introduce the sequence attributes for the performance analysis. Second, we integrate most of the publicly available trackers into one code library with uniform input and output formats to facilitate large-scale performance evaluation. Third, we extensively evaluate the performance of 31 algorithms on 100 sequences with different initialization settings. By analyzing the quantitative results, we identify effective approaches for robust tracking and provide potential future research directions in this field.
2,974 citations
Cites background or methods from "Color-Based Probabilistic Tracking"
...Mei and Ling [53], [54] used a dictionary of holistic intensity templates composed of target and trivial templates, and determined the target location by solving multiple ‘1 minimization problems....
[...]
...Here, we discuss the relevant performance evaluation work on object tracking and challenging factors in object tracking....
[...]
TL;DR: The proposed SRDCF formulation allows the correlation filters to be learned on a significantly larger set of negative training samples, without corrupting the positive samples, and an optimization strategy is proposed, based on the iterative Gauss-Seidel method, for efficient online learning.
Abstract: Robust and accurate visual tracking is one of the most challenging computer vision problems. Due to the inherent lack of training data, a robust approach for constructing a target appearance model is crucial. Recently, discriminatively learned correlation filters (DCF) have been successfully applied to address this problem for tracking. These methods utilize a periodic assumption of the training samples to efficiently learn a classifier on all patches in the target neighborhood. However, the periodic assumption also introduces unwanted boundary effects, which severely degrade the quality of the tracking model.
We propose Spatially Regularized Discriminative Correlation Filters (SRDCF) for tracking. A spatial regularization component is introduced in the learning to penalize correlation filter coefficients depending on their spatial location. Our SRDCF formulation allows the correlation filters to be learned on a significantly larger set of negative training samples, without corrupting the positive samples. We further propose an optimization strategy, based on the iterative Gauss-Seidel method, for efficient online learning of our SRDCF. Experiments are performed on four benchmark datasets: OTB-2013, ALOV++, OTB-2015, and VOT2014. Our approach achieves state-of-the-art results on all four datasets. On OTB-2013 and OTB-2015, we obtain an absolute gain of 8.0% and 8.2% respectively, in mean overlap precision, compared to the best existing trackers.
1,616 citations
Cites methods from "Color-Based Probabilistic Tracking"
...We provide a comparison of our tracker with 24 state-ofthe-art methods from the literature: MIL [2], IVT [28], CT [36], TLD [22], DFT [29], EDFT [12], ASLA [21], L1APG [3], CSK [19], SCM [37], LOT [26], CPF [27], CXT [11], Frag [1], Struck [16], LSHT [17], LSST [32], ACT [10], KCF [20], CFLB [14], DSST [8], SAMF [24], TGPR [15] and MEEM [35]....
[...]
TL;DR: A framework for adaptive visual object tracking based on structured output prediction that is able to outperform state-of-the-art trackers on various benchmark videos and can easily incorporate additional features and kernels into the framework, which results in increased tracking performance.
Abstract: Adaptive tracking-by-detection methods are widely used in computer vision for tracking arbitrary objects. Current approaches treat the tracking problem as a classification task and use online learning techniques to update the object model. However, for these updates to happen one needs to convert the estimated object position into a set of labelled training examples, and it is not clear how best to perform this intermediate step. Furthermore, the objective for the classifier (label prediction) is not explicitly coupled to the objective for the tracker (estimation of object position). In this paper, we present a framework for adaptive visual object tracking based on structured output prediction. By explicitly allowing the output space to express the needs of the tracker, we avoid the need for an intermediate classification step. Our method uses a kernelised structured output support vector machine (SVM), which is learned online to provide adaptive tracking. To allow our tracker to run at high frame rates, we (a) introduce a budgeting mechanism that prevents the unbounded growth in the number of support vectors that would otherwise occur during tracking, and (b) show how to implement tracking on the GPU. Experimentally, we show that our algorithm is able to outperform state-of-the-art trackers on various benchmark videos. Additionally, we show that we can easily incorporate additional features and kernels into our framework, which results in increased tracking performance.
1,507 citations
Cites methods from "Color-Based Probabilistic Tracking"
...The trackers used included MILTrack [3], MTT [46] and ColorPF [47]....
[...]
References
More filters
TL;DR: The Condensation algorithm uses “factored sampling”, previously applied to the interpretation of static images, in which the probability distribution of possible interpretations is represented by a randomly generated set.
Abstract: The problem of tracking curves in dense visual clutter is challenging. Kalman filtering is inadequate because it is based on Gaussian densities which, being unimo dal, cannot represent simultaneous alternative hypotheses. The Condensation algorithm uses “factored sampling”, previously applied to the interpretation of static images, in which the probability distribution of possible interpretations is represented by a randomly generated set. Condensation uses learned dynamical models, together with visual observations, to propagate the random set over time. The result is highly robust tracking of agile motion. Notwithstanding the use of stochastic methods, the algorithm runs in near real-time.
5,804 citations
"Color-Based Probabilistic Tracking" refers methods in this paper
...Sequential Monte Carlo techniques for filtering time series [6] and their use in the specific context of visual tracking [10] have been described at length in the literature....
[...]
..., an ellipse outline for faces [1,18], or extracted from a set of examples like gray level appearance templates in [2] or outlines in [8,10,12]....
[...]
Book•
01 Jan 1995
TL;DR: This chapter discusses the development of Hardware and Software for Computer Graphics, and the design methodology of User-Computer Dialogues, which led to the creation of the Simple Raster Graphics Package.
Abstract: 1 Introduction Image Processing as Picture Analysis The Advantages of Interactive Graphics Representative Uses of Computer Graphics Classification of Applications Development of Hardware and Software for Computer Graphics Conceptual Framework for Interactive Graphics 2 Programming in the Simple Raster Graphics Package (SRGP)/ Drawing with SRGP/ Basic Interaction Handling/ Raster Graphics Features/ Limitations of SRGP/ 3 Basic Raster Graphics Algorithms for Drawing 2d Primitives Overview Scan Converting Lines Scan Converting Circles Scan Convertiing Ellipses Filling Rectangles Fillign Polygons Filling Ellipse Arcs Pattern Filling Thick Primiives Line Style and Pen Style Clipping in a Raster World Clipping Lines Clipping Circles and Ellipses Clipping Polygons Generating Characters SRGP_copyPixel Antialiasing 4 Graphics Hardware Hardcopy Technologies Display Technologies Raster-Scan Display Systems The Video Controller Random-Scan Display Processor Input Devices for Operator Interaction Image Scanners 5 Geometrical Transformations 2D Transformations Homogeneous Coordinates and Matrix Representation of 2D Transformations Composition of 2D Transformations The Window-to-Viewport Transformation Efficiency Matrix Representation of 3D Transformations Composition of 3D Transformations Transformations as a Change in Coordinate System 6 Viewing in 3D Projections Specifying an Arbitrary 3D View Examples of 3D Viewing The Mathematics of Planar Geometric Projections Implementing Planar Geometric Projections Coordinate Systems 7 Object Hierarchy and Simple PHIGS (SPHIGS) Geometric Modeling Characteristics of Retained-Mode Graphics Packages Defining and Displaying Structures Modeling Transformations Hierarchical Structure Networks Matrix Composition in Display Traversal Appearance-Attribute Handling in Hierarchy Screen Updating and Rendering Modes Structure Network Editing for Dynamic Effects Interaction Additional Output Features Implementation Issues Optimizing Display of Hierarchical Models Limitations of Hierarchical Modeling in PHIGS Alternative Forms of Hierarchical Modeling 8 Input Devices, Interaction Techniques, and Interaction Tasks Interaction Hardware Basic Interaction Tasks Composite Interaction Tasks 9 Dialogue Design The Form and Content of User-Computer Dialogues User-Interfaces Styles Important Design Considerations Modes and Syntax Visual Design The Design Methodology 10 User Interface Software Basic Interaction-Handling Models Windows-Management Systems Output Handling in Window Systems Input Handling in Window Systems Interaction-Technique Toolkits User-Interface Management Systems 11 Representing Curves and Surfaces Polygon Meshes Parametric Cubic Curves Parametric Bicubic Surfaces Quadric Surfaces 12 Solid Modeling Representing Solids Regularized Boolean Set Operations Primitive Instancing Sweep Representations Boundary Representations Spatial-Partitioning Representations Constructive Solid Geometry Comparison of Representations User Interfaces for Solid Modeling 13 Achromatic and Colored Light Achromatic Light Chromatic Color Color Models for Raster Graphics Reproducing Color Using Color in Computer Graphics 14 The Quest for Visual Realism Why Realism? Fundamental Difficulties Rendering Techniques for Line Drawings Rendering Techniques for Shaded Images Improved Object Models Dynamics Stereopsis Improved Displays Interacting with Our Other Senses Aliasing and Antialiasing 15 Visible-Surface Determination Functions of Two Variables Techniques for Efficient Visible-Surface Determination Algorithms for Visible-Line Determination The z-Buffer Algorithm List-Priority Algorithms Scan-Line Algorithms Area-Subdivision Algorithms Algorithms for Octrees Algorithms for Curved Surfaces Visible-Surface Ray Tracing 16 Illumination And Shading Illumination Modeling Shading Models for Polygons Surface Detail Shadows Transparency Interobject Reflections Physically Based Illumination Models Extended Light Sources Spectral Sampling Improving the Camera Model Global Illumination Algorithms Recursive Ray Tracing Radiosity Methods The Rendering Pipeline 17 Image Manipulation and Storage What Is an Image? Filtering Image Processing Geometric Transformations of Images Multipass Transformations Image Compositing Mechanisms for Image Storage Special Effects with Images Summary 18 Advanced Raster Graphic Architecture Simple Raster-Display System Display-Processor Systems Standard Graphics Pipeline Introduction to Multiprocessing Pipeline Front-End Architecture Parallel Front-End Architectures Multiprocessor Rasterization Architectures Image-Parallel Rasterization Object-Parallel Rasterization Hybrid-Parallel Rasterization Enhanced Display Capabilities 19 Advanced Geometric and Raster Algorithms Clipping Scan-Converting Primitives Antialiasing The Special Problems of Text Filling Algorithms Making copyPixel Fast The Shape Data Structure and Shape Algebra Managing Windows with bitBlt Page Description Languages 20 Advanced Modeling Techniques Extensions of Previous Techniques Procedural Models Fractal Models Grammar-Based Models Particle Systems Volume Rendering Physically Based Modeling Special Models for Natural and Synthetic Objects Automating Object Placement 21 Animation Conventional and Computer-Assisted Animation Animation Languages Methods of Controlling Animation Basic Rules of Animation Problems Peculiar to Animation Appendix: Mathematics for Computer Graphics Vector Spaces and Affine Spaces Some Standard Constructions in Vector Spaces Dot Products and Distances Matrices Linear and Affine Transformations Eigenvalues and Eigenvectors Newton-Raphson Iteration for Root Finding Bibliography Index 0201848406T04062001
5,692 citations
TL;DR: An overview of methods for sequential simulation from posterior distributions for discrete time dynamic models that are typically nonlinear and non-Gaussian, and how to incorporate local linearisation methods similar to those which have previously been employed in the deterministic filtering literature are shown.
Abstract: In this article, we present an overview of methods for sequential simulation from posterior distributions. These methods are of particular interest in Bayesian filtering for discrete time dynamic models that are typically nonlinear and non-Gaussian. A general importance sampling framework is developed that unifies many of the methods which have been proposed over the last few decades in several different scientific disciplines. Novel extensions to the existing methods are also proposed. We show in particular how to incorporate local linearisation methods similar to those which have previously been employed in the deterministic filtering literatures these lead to very effective importance distributions. Furthermore we describe a method which uses Rao-Blackwellisation in order to take advantage of the analytic structure present in some important classes of state-space models. In a final section we develop algorithms for prediction, smoothing and evaluation of the likelihood in dynamic models.
4,810 citations
"Color-Based Probabilistic Tracking" refers background or methods in this paper
...Sequential Monte Carlo techniques for filtering time series [6] and their use in the specific context of visual tracking [10] have been described at length in the literature....
[...]
...It can be shown (see [6]) that the optimal proposal density is proportional to p(yt+1| x̃t+1)p(x̃t+1|xt), but its normalization ∫ x̃t+1 p(yt+1| x̃t+1)p(x̃t+1|xt) cannot be computed analytically in our case....
[...]
3,375 citations
14 Feb 2000
TL;DR: The theoretical analysis of the approach shows that it relates to the Bayesian framework while providing a practical, fast and efficient solution for real time tracking of non-rigid objects seen from a moving camera.
Abstract: A new method for real time tracking of non-rigid objects seen from a moving camera is proposed. The central computational module is based on the mean shift iterations and finds the most probable target position in the current frame. The dissimilarity between the target model (its color distribution) and the target candidates is expressed by a metric derived from the Bhattacharyya coefficient. The theoretical analysis of the approach shows that it relates to the Bayesian framework while providing a practical, fast and efficient solution. The capability of the tracker to handle in real time partial occlusions, significant clutter, and target scale variations, is demonstrated for several image sequences.
3,368 citations
"Color-Based Probabilistic Tracking" refers background or methods in this paper
...(“MeanShift” in [5]), and modified later by Chen et al....
[...]
...It can be an ellipse or a rectangular box as in [3,4,5]....
[...]
...In [3,4,5] the weight function is a smooth kernel such that the gradient computations required by the iterative optimization process can be performed....
[...]
...Color-based trackers recently proposed in [3,4,5] have been proved robust and versatile for a modest computational cost....
[...]
...In that case the reference can be extracted from the first frame and kept frozen, as color models in [3,4,5] and gray-level templates in [9], or adapted on the fly, using the tracking results from the previous frames, as gray-level templates in [14,15,17], deformable outlines in [16], and color models in [18,21,20]....
[...]