Worthy N. Martin

Human-computer interaction using eye-gaze input

A description is given of Erica, a computer workstation with a unique user interface. The workstation is equipped with imaging hardware and software, which automatically record a digital portrait of the users eye. From the features of the current portrait, the interface calculates the approximate location of the users eye-gaze on the computer screen. The computer then executes commands associated with the menu option currently displayed at this screen location. In this way, the user can interact with the computer, run applications software, and manage peripheral devices-all simply by looking at an appropriate sequence of menu options displayed on the screen. The eye-gaze interface technology, its implementation in Erica, and its application as a prosthetic device are described. >

Fast generic selection of features for neural network classifiers

The authors describe experiments using a genetic algorithm for feature selection in the context of neural network classifiers, specifically, counterpropagation networks. They present the novel techniques used in the application of genetic algorithms. First, the genetic algorithm is configured to use an approximate evaluation in order to reduce significantly the computation required. In particular, though the desired classifiers are counterpropagation networks, they use a nearest-neighbor classifier to evaluate features sets and show that the features selected by this method are effective in the context of counterpropagation networks. Second, a method called the training set sampling in which only a portion of the training set is used on any given evaluation, is proposed. Computational savings can be made using this method, i.e., evaluations can be made over an order of magnitude faster. This method selects feature sets that are as good as and occasionally better for counterpropagation than those chosen by an evaluation that uses the entire training set.

Distributed genetic algorithms for the floorplan design problem

Designing a VLSI floorplan calls for arranging a given set of modules in the plane to minimize the weighted sum of area and wire-length measures. A method of solving the floorplan design problem using distributed genetic algorithms is presented. Distributed genetic algorithms, based on the paleontological theory of punctuated equilibria, offer a conceptual modification to the traditional genetic algorithms. Experimental results on several problem instances demonstrate the efficacy of this method and indicate the advantages of this method over other methods, such as simulated annealing. The method has performed better than the simulated annealing approach, both in terms of the average cost of the solutions found and the best-found solution, in almost all the problem instances tried. >

Correspondence processes in dynamic scene analysis

One of the fundamental problems in dynamic scene analysis is the tracking of objects from frame to frame. A general approach to tracking is to establish correspondences between points, or sets of points, between frames and then group the sets into objects based upon similarity of motion. This paper will focus on processes for establishing the correspondence between sets of points in successive frames. A succession of correspondence processes are discussed, based on the factors which contribute to the complexity of the correspondence problem.

Segmentation through the detection of changes due to motion

Abstract This paper discusses a scheme for extracting the images of moving objects in dynamic scenes. Differencing operations are used to identify areas containing moving objects. The images of the moving objects can then be obtained by focusing the segmentation processes on these restricted areas. Thus motion is used as a cue to segmentation.

Dynamic scene analysis

The three major components of dynamic scene analysis, namely segmentation, occlusion and the computation of three-dimensional information from images are discussed in depth. Segmentation refers to the process of determining features of interest, occlusion analysis includes the deriving of changes due to projection perspective, and computation of three-dimensional information entails the constructing of structural models and describing motions from image information. The purpose of the review is to give the reader a coherent view of the issues and the manner in which researchers are currently addressing these issues. Detailed descriptions of the systems developed at The University of Texas are presented.

Using the triangle inequality to reduce the number of comparisons required for similarity-based retrieval

Dissimilarity measures, the basis of similarity-based retrieval, can be viewed as a distance and a similarity-based search as a nearest neighbor search. Though there has been extensive research on data structures and search methods to support nearest-neighbor searching, these indexing and dimension-reduction methods are generally not applicable to non-coordinate data and non-Euclidean distance measures. In this paper we reexamine and extend previous work of other researchers on best match searching based on the triangle inequality. These methods can be used to organize both non-coordinate data and non-Euclidean metric similarity measures. The effectiveness of the indexes depends on the actual dimensionality of the feature set, data, and similarity metric used. We show that these methods provide significant performance improvements and may be of practical value in real-world databases.

Computer analysis of dynamic scenes containing curvilinear figures

Abstract A brief introduction to dynamic scene analysis is given, followed by a description of a system which analyzes sequences of images of moving figures. The methods used to describe the boundaries of the figures, to match common shapes between paired images, and then to analyze the motion of matched figures, are discussed in detail. Finally, a complete example is presented along with some directions for further research.

Extraction of moving object descriptions via differencing

Abstract In the analysis of scenes containing objects in motion a primary concern is the ability to identify and derive descriptions of moving objects. The motion of these objects introduces changes in the images of the scene. A system has been developed which detects changes due to motion and examines these changes to obtain samples of the objects. A “region growing” process uses these samples to derive image plane descriptions of the moving objects in the scene.

Genetic Algorithms and Punctuated Equilibria in VLSI

The distributed genetic algorithm presented has a population structure that allows the introduction of “ecological opportunity” [Wrig 82] in the evolutionary process in a manner motivated by the macro-evolutionary theory of Eldredge and Gould [Eldr 72]. The K-partition problem is selected from the domain of VLSI design and empirical results are presented to show the advantage derived from the population structure.