AbstractÐWe describe a new method of matching statistical models of appearance to images. A set of model parameters control modes of shape and gray-level variation learned from a training set. We construct an efficient iterative matching algorithm by learning the relationship between perturbations in the model parameters and the induced image errors. Index TermsÐAppearance models, deformable templates, model matching. 1

Since the Lucas-Kanade algorithm was proposed in 1981 image alignment has become one of the most widely used techniques in computer vision. Applications range from optical flow, tracking, and layered motion, to mosaic construction, medical image registration, and face coding. Numerous algorithms have been proposed and a variety of extensions have been made to the original formulation. We present an overview of image alignment, describing most of the algorithms in a consistent framework. We concentrate on the inverse compositional algorithm, an efficient algorithm that we recently proposed. We examine which of the extensions to the Lucas-Kanade algorithm can be used with the inverse compositional algorithm without any significant loss of efficiency, and which cannot. In this paper, Part 3 in a series of papers, we cover the extension of image alignment to allow linear appearance variation. We first consider linear appearance variation when the error function is the Euclidean L2 norm. We describe three different algorithms, the simultaneous, project out, and normalization inverse compositional algorithms, and empirically compare them. Afterwards we consider the combination of linear appearance variation with the robust error functions described in Part 2 of this series. We first derive robust versions of the simultaneous and normalization algorithms. Since both of these algorithms are very inefficient, as in Part 2 we derive efficient approximations based on spatial coherence. We end with an empirical evaluation of the robust algorithms.

Active Appearance Models (AAMs) and the closely related concepts of Morphable Models and Active Blobs are generative models of a certain visual phenomenon. Although linear in both shape and appearance, overall, AAMs are nonlinear parametric models in terms of the pixel intensities. Fitting an AAM to an image consists of minimizing the error between the input image and the closest model instance; i.e. solving a nonlinear optimization problem. We propose an efficient fitting algorithm for AAMs based on the inverse compositional image alignment algorithm. We show how the appearance variation can be "projected out" using this algorithm and how the algorithm can be extended to include a "shape normalizing" warp, typically a 2D similarity transformation. We evaluate our algorithm to determine which of its novel aspects improve AAM fitting performance.

Abstract not found

We propose a framework for learning robust, adaptive, appearance models to be used for motion-based tracking of natural objects. The approach involves a mixture of stable image structure, learned over long time courses, along with 2-frame motion information and an outlier process. An online EM-algorithm is used to adapt the appearance model parameters over time. An implementation of this approach is developed for an appearance model based on the filter responses from a steerable pyramid. This model is used in a motion-based tracking algorithm to provide robustness in the face of image outliers, such as those caused by occlusions. It is also provides the ability to adapt to natural changes in appearance, such as those due to facial expressions or variations in 3D pose. We show experimental results on a variety of natural image sequences of people moving within cluttered environments.

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 model. Relying 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 frames. This 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. 1

This paper presents an overview of the issues and algorithms involved in creating this semiautonomous, multicamera surveillance system

There are two major formulations of image alignment using gradient descent. The first estimates an additive increment to the parameters (the additive approach), the second an incremental warp (the compositional approach). We first prove that these two formulations are equivalent. A very efficient algorithm was recently proposed by Hager and Belhumeur using the additive approach that unfortunately can only be applied to a very restricted class of warps. We show that using the compositional approach an equally efficient algorithm (the inverse compositional algorithm) can be derived that can be applied to any set of warps which form a group. While most warps used in computer vision form groups, there are a certain warps that do not. Perhaps most notable is the set of piecewise affine warps used in Flexible Appearance Models (FAMs). We end this paper by extending the inverse compositional algorithm to apply to FAMs. 1

Text which appears in a scene or is graphically added to video can provide an important supplemental source of index information as well as clues for decoding the video's structure and for classification. In this paper we present algorithms for detecting and tracking text in digital video. Our system implements a scalespace feature extractor that feeds an artificial neural processor to detect text blocks. Our text tracking scheme consists of two modules: an SSD (Sum of Squared Difference)-based module to find the initial position and a contour-based module to refine the position. Experiments conducted with a variety of video sources show that our scheme can detect and track text robustly.

Template tracking is a well studied problem in computer vision which dates back to the Lucas-Kanade algorithm of 1981. Since then the paradigm has been extended in a variety of ways including: arbitrary parametric transformations of the template, and linear appearance variation. These extensions have been combined, culminating in non-rigid appearance models such as Active Appearance Models (AAMs) and Active Blobs. One question that has received very little attention is how to update the template over time so that it remains a good model of the object being tracked. This paper proposes an algorithm to update the template that avoids the "drifting" problem of the naive update algorithm. Our algorithm can be interpreted as a heuristic to avoid local minima. It can also be extended to templates with linear appearance variation. This extension can be used to convert (update) a generic, person-independent AAM into a person specific AAM.