Publications (11)11.07 Total impact
 [Show abstract] [Hide abstract]
ABSTRACT: The mainstream object categorisation community relies heavily on object representations consisting of local image features, due to their ease of recovery and their attractive invariance properties. Object categorisation is therefore formulated as finding, that is, `detecting`, a onetoone correspondence between image and model features. This assumption breaks down for categories in which two exemplars may not share a single local image feature. Even when objects are represented as more abstract image features, a collection of features at one scale (in one image) may correspond to a single feature at a coarser scale (in the second image). Effective object categorisation therefore requires the ability to match features manytomany. In this paper, we review our progress on three independent object categorisation problems, each formulated as a graph matching problem and each solving the manytomany graph matching problem in a different way. First, we explore the problem of learning a shape class prototype from a set of class exemplars which may not share a single local image feature. Next, we explore the problem of matching two graphs in which correspondence exists only at higher levels of abstraction, and describe a lowdimensional, spectral encoding of graph structure that captures the abstract shape of a graph. Finally, we embed graphs into geometric spaces, reducing the manytomany graphmatching problem to a weighted point matching problem, for which efficient manytomany matching algorithms exist.  [Show abstract] [Hide abstract]
ABSTRACT: Object recognition can be formulated as matching image features to model features. When recognition is exemplarbased, feature correspondence is onetoone. However, segmentation errors, articulation, scale difference, and withinclass deformation can yield image and model features which don’t match onetoone but rather manytomany. Adopting a graphbased representation of a set of features, we present a matching algorithm that establishes manytomany correspondences between the nodes of two noisy, vertexlabeled weighted graphs. Our approach reduces the problem of manytomany matching of weighted graphs to that of manytomany matching of weighted point sets in a normed vector space. This is accomplished by embedding the initial weighted graphs into a normed vector space with low distortion using a novel embedding technique based on a spherical encoding of graph structure. Manytomany vector correspondences established by the Earth Mover’s Distance framework are mapped back into manytomany correspondences between graph nodes. Empirical evaluation of the algorithm on an extensive set of recognition trials, including a comparison with two competing graph matching approaches, demonstrates both the robustness and efficacy of the overall approach. 
Conference Paper: ManytoMany Feature Matching in Object Recognition.
[Show abstract] [Hide abstract]
ABSTRACT: One of the bottlenecks of current recognition (and graph matching) systems is their assumption of onetoone feature (node) correspondence. This assumption breaks down in the generic object recognition task where, for example, a collection of features at one scale (in one image) may correspond to a single feature at a coarser scale (in the second image). Generic object recognition therefore requires the ability to match features manytomany. In this paper, we will review our progress on three independent object recognition problems, each formulated as a graph matching problem and each solving the manytomany matching problem in a different way. First, we explore the problem of learning a 2D shape class prototype (represented as a graph) from a set of object exemplars (also represented as graphs) belonging to the class, in which there may be no onetoone correspondence among extracted features. Next, we define a lowdimensional, spectral encoding of graph structure and use it to match entire subgraphs whose size can be different. Finally, in very recent work, we embed graphs into geometric spaces, reducing the manytomany graph matching problem to a weighted point matching problem, for which efficient manytomany matching algorithms exist.  [Show abstract] [Hide abstract]
ABSTRACT: The recognition community has typically avoided bridging the representational gap between traditional, lowlevel image features and generic models. Instead, the gap has been artificially eliminated by either bringing the image closer to the models using simple scenes containing idealized, textureless objects or by bringing the models closer to the images using 3D CAD model templates or 2D appearance model templates. In this paper, we attempt to bridge the representational gap for the domain of model acquisition. Specifically, we address the problem of automatically acquiring a generic 2D viewbased class model from a set of images, each containing an exemplar object belonging to that class. We introduce a novel graphtheoretical formulation of the problem in which we search for the lowest common abstraction among a set of lattices, each representing the space of all possible region groupings in a region adjacency graph representation of an input image. The problem is intractable and we present a shortest pathbased approximation algorithm to yield an efficient solution. We demonstrate the approach on real imagery.  [Show abstract] [Hide abstract]
ABSTRACT: Object recognition systems have their roots in the AI com munity, and originally addressed the problem of object categorization. These early systems, however, were limited by their inability to bridge the representational gap between lowlevel image features and highlevel object models, hindered by the assumption of onetoone correspondence between image and model features. Over the next thirty years, the main stream recognition community moved steadily in the direction of exem plar recognition while narrowing the representational gap. The commu nity is now returning to the categorization problem, and faces the same representational gap as its predecessors did. We review the evolution of object recognition systems and argue that bridging this representa tional gap requires an ability to match image and model features many tomany. We review three formulations of the manytomany matching problem as applied to model acquisition and object recognition. 
Article: ManytoMany Feature Matching Using
[Show abstract] [Hide abstract]
ABSTRACT: In recent work, we presented a framework for manytomany matching of multiscale feature hierarchies, in which features and their relations were captured in a vertexlabeled, edgeweighted directed graph. 
Conference Paper: ManytoMany Feature Matching Using Spherical Coding of Directed Graphs
[Show abstract] [Hide abstract]
ABSTRACT: In recent work, we presented a framework for manytomany matching of multiscale feature hierarchies, in which features and their re lations were captured in a vertexlabeled, edgeweighted directed graph. The algorithm was based on a metrictree representation of labeled graphs and their metric embedding into normed vector spaces, using the embed ding algorithm of Matou sek (13). However, the method was limited by the fact that two graphs to be matched were typically embedded into vector spaces with dieren t dimensionality. Before the embeddings could be matched, a dimensionality reduction technique (PCA) was required, which was both costly and prone to error. In this paper, we introduce a more ecien t embedding procedure based on a spherical coding of di rected graphs. The advantage of this novel embedding technique is that it prescribes a single vector space into which both graphs are embedded. This reduces the problem of directed graph matching to the problem of geometric point matching, for which ecien t manytomany match ing algorithms exist, such as the Earth Mover's Distance. We apply the approach to the problem of multiscale, viewbased object recognition, in which an image is decomposed into a set of blobs and ridges with automatic scale selection. 
Conference Paper: Manytomany graph matching via metric embedding
[Show abstract] [Hide abstract]
ABSTRACT: Graph matching is an important component in many object recognition algorithms. Although most graph matching algorithms seek a onetoone correspondence between nodes, it is often the case that a more meaningful correspondence exists between a cluster of nodes in one graph and a cluster of nodes in the other. We present a matching algorithm that establishes manytomany correspondences between nodes of noisy, vertexlabeled weighted graphs. The algorithm is based on recent developments in efficient lowdistortion metric embedding of graphs into normed vector spaces. By embedding weighted graphs into normed vector spaces, we reduce the problem of manytomany graph matching to that of computing a distributionbased distance measure between graph embeddings. We use a specific measure, the earth mover's distance, to compute distances between sets of weighted vectors. Empirical evaluation of the algorithm on an extensive set of recognition trials demonstrates both the robustness and efficiency of the overall approach.  [Show abstract] [Hide abstract]
ABSTRACT: Scalespace feature hierarchies can be conveniently represented as graphs, in which edges are directed from coarser features to finer features. Consequently, feature matching (or viewbased object matching) can be formulated as graph matching. Most approaches to graph matching assume a onetoone correspondence between nodes (features) which, due to noise, scale discretization, and feature extraction errors, is overly restrictive. In general, a subset of features in one hierarchy, representing an abstraction of those features, may best match a subset of features in another. We present a framework for the manytomany matching of multiscale feature hierarchies, in which features and their relations are captured in a vertexlabeled, edgeweighted graph. The matching algorithm is based on a metrictree representation of labeled graphs and their lowdistortion metric embedding into normed vector spaces. This twostep transformation reduces the manytomany graph matching problem to that of computing a distributionbased distance measure between two such embeddings. To compute the distance between two sets of embedded, weighted vectors, we use the Earth Mover's Distance under transformation. To demonstrate the approach, we target the domain of multiscale, qualitative shape description, in which an image is decomposed into a set of blobs and ridges with automatic scale selection. We conduct an extensive set of viewbased matching trials, and compare the results favorably to matching under a onetoone assumption.  [Show abstract] [Hide abstract]
ABSTRACT: The recognition community has long avoided bridging the representational gap between traditional, lowlevel image features and generic models. Instead, the gap has been ar tificially eliminated by either bringing the image closer to the models, using simple scenes containing idealized, tex tureless objects, or by bringing the models closer to the im ages, using 3D CAD model templates or 2D appearance model templates. In this paper, we begin by examining this trend and track its evolution over the last 30 years. We ar gue for the need to bridge (not eliminate) this representa tional gap, and review our recent progress for the domain of model acquisition. Specifically, we address the problem of automatically acquiring a generic 2D viewbased class model from a set of images, each containing an exemplar object belonging to that class. We introduce a novel graph theoreticalformulationof the problem, anddemonstratethe approach on real imagery. 
Article: Evolving ANN for Edge Detection
[Show abstract] [Hide abstract]
ABSTRACT: The authors investigate the use of a genetic algorithm to control the evolution of artificial neural networks for the purpose of detecting edges in singleline digitized images. Specialpurpose neurons are used in an architecture which is fixed in terms of interlayer connectivity but free in terms of number of layers, activation functions, and other attributes. Preliminary results indicate that three of the free attributes do have a best form.
Publication Stats
332  Citations  
11.07  Total Impact Points  
Top Journals
Institutions

20032006

DePaul University
Chicago, Illinois, United States
