Louise Stark

Generic recognition through qualitative reasoning about 3-D shape and object function

The work which demonstrates the feasibility of a different approach to 3-D object recognition is described. The authors construct a definition of a generic object category, such as a chair, in terms of the function required of the object. This definition is based on qualitative reasoning about 3-D shape, and does not imply any particular geometric or structural model for an object. Thus, this approach has the potential to lead to recognition systems of much greater generality than current CAD-based or model-based approaches. >

Aspect Graphs And Nonlinear Optimization In 3-D Object Recognition

Several researchers have previously described approaches to 3-D object recognition which use nonlinear optimization to control the matching of features of a 3-D object niodel to features found in an image. Recognition, in this context, includes estimating the parameters of translation and orientation of the object. A major problem acknowledged by previous researchers is how to efficiently choose a set of starting paranleter estimates which will avoid recognition errors due to local minima. The unique contribution of this paper is that it outlines an approach for using the perspective projection aspect gruph representation to alleviate the problems encountered by previous researchers, describes a particular implementation of this general approach, and presents data to illustrate the effectiveness (of the approac!].

GRUFF-3: Generalizing the domain of a function-based recognition system☆

Abstract Representation systems which support “generic” object recognition offer promising advantages over current model-based vision. Systems applying function-based reasoning are one such approach. In this approach, specific geometric or structural models are disregarded, in favor of analyzing the shape to determine functional requirements for category membership. This paper presents an explanation of the ideas behind function-based modeling and a description of the extensions made to create the Generic Representation Using Form and Function-3 (GRUFF-3) system. This system analyzes the 3D shape of an object and classifies the object according to its possible function as some (sub) category of the superordinate category dishes . The initial GRUFF system implementation was restricted to the furniture domain and required five knowledge primitives (clearance, relative orientation, proximity, dimensions and stability) to realize the functional requirements of the categories represented. The important contribution of our current work is that a significantly larger domain of objects can now be recognized with the addition of just one new knowledge primitive, enclosure. An evaluation of the performance of the system is presented for a database of over 200 3D shapes.

Human perceptual categorization of iris texture patterns

We report on an experiment in which observers were asked to browse a set of 100 iris images and group them into categories based on similarity of overall texture appearance. Results indicate that there is a natural categorization of iris images into a small number of high-level categories, and then also into subcategories. Also, the categorization reflects the Caucasian / Asian ethnicity of the person. This iris texture categorization has potential application in, for example, creating an indexing algorithm to speed search of an iris database and / or determining soft biometric traits of a person.

Themes for improved teaching of image computation

This paper reports on recommendations and results from a panel discussion and a workshop devoted to the theme of education in areas that involve image computation. One specific set of contributions is in the area of integrating image computation into required core courses such as Introduction to Computing and Data Structures. Another area of specific contributions is the development of undergraduate elective courses on topics such as robotics and medical image analysis. A third area of contributions is the improvement of traditional image processing and computer vision courses.

Indexing function-based categories for generic recognition

The authors report the implementation and evaluation of a function-based recognition system that takes an uninterrupted 3-D object shape as its input and reasons to determine if the object belongs to the superordinate category furniture, and if so, into which (sub)category it falls. The system has analyzed over 250 input objects, and the results largely agree with intuitive human interpretation of the objects. The study confirms that a relatively small number of knowledge primitives may be used as the basis for defining a relatively broad range of object categories. The greatest derivation from intuitive human interpretation occurs with objects that humans would not typically label as one of the known categories defined, but which have some novel orientation in which they could serve the function of one of these categories. This is because the system uses a purely function-based definition of the object category. >

Exploiting Context in Function-Based Reasoning

This paper presents the framework of the new context-based reasoning components of the GRUFF (Generic Recognition Using Form and Function) system. This is a generic object recognition system which reasons about and generates plans for understanding 3-D scenes of objects. A range image is generated from a stereo image pair and is provided as input to a multi-stage recognition system. A 3-D model of the scene, extracted from the range image, is processed to identify evidence of potential functionality directed by contextual cues. This recognition process considers the shape-suggested functionality by applying concepts of physics and causation to label an objects potential functionality. The methodology for context-based reasoning relies on determining the significance of the accumulated functional evidence derived from the scene. For example, functional evidence for a chair or multiple chairs along with a table, in set configurations, is used to infer the existence of scene concepts such as office or meeting room space. Results of this work are presented for scene understanding derived from both simulated and real sensors positioned in typical office and meeting room environments.

Aspect graphs and their use in object recognition

Previous researchers have described several different approaches to 3-D object recognition based on using an iterative technique to control the matching of features from the 2-D projection of a 3-D model to observed image features. The major problem encountered with such approaches is how to automatically choose starting parameter estimates in a manner which both avoids recognition errors due to local minima and is still reasonably efficient. This paper investigates the use of theaspect graph to address this problem. The basic idea is quite simple — an iterative solution is generated for each of a set of candidate aspects and the best of these is chosen as the recognized view. Two assumptions are required in order for this approach to be valid: (1) the iterative search for the correct candidate aspect must converge to the correct answer, and (2) the solution found for the correct aspect must be better than that found for any of the incorrect candidate aspects. In order to explore the validity of these assumptions, a simple aspect graph-based recognition system was implemented. Experiments were carried out using both real and simulated data. The results indicate that the underlying assumptions are generally valid, and that this approach has advantages over previous techniques which incorporated an iterative search.

Generic recognition of articulated objects by reasoning about functionality

Previous work on the recognition of objects by reasoning about their functionality has not dealt with objects that have moving parts. In this paper we introduce a scenario in which object recognition is accomplished by first deriving an articulated shape model from an observed sequence of 3-D shapes and by then reasoning about the possible functionality of the articulated shape model.

METHODS FOR COMBINATION OF EVIDENCE IN FUNCTION-BASED 3-D OBJECT RECOGNITION

Representation schemes traditionally used in model-based vision are contrasted with the “function-based” representation scheme. A system which utilizes function-based representation has been implemented and tested, using the object category “chair” for case study. Function-based description is used to recognize classes and identify subclasses of known categories of objects, even if the specific object has never been encountered previously. Interpretation of the functionality of an object is accomplished through qualitative reasoning about its 3-D shape. During the recognition process, evidence is gathered as to how well the functional requirements are satisfied by the input shape. An investigation of different types of operators used in the combination of the functional evidence has been made. Three pairs of conjunctive and disjunctive operators have been used in the recognition process of more than 100 object shapes. The results are compared and differences are discussed.