Dudek, G., Jenkin, M., and Marcus, H., How to make friends with number-crunchers: adding single-user array-processor slave environments to VAX UNIX. Proc. 1986 Summer USENIX Conference, 200-208, Atlanta, GA, 1986.
There is a growing need for array processors or dedicated machines for scientific applications. In particular, computer vision and computer graphics are computationally intensive and require sophisticated display hardware. Unfortunately, the current state of the art makes it difficult to develop software for such devices, and few, if any, full development environments exist for UNIX today.This paper describes the design and implementation of a transparent development environment for an array-processor and graphics subsystem. A number of critical issues are addressed in this paper; in particular, the problem of transparently providing remote CPU access and of allowing a task running on a slave machine to access the host without slave-specific operations are addressed.
Jenkin, M., and Tsotsos, J. K., Applying temporal constraints to the dynamic stereo problem, Computer Vision, Graphics and Image Processing, 33: 16-32, 1986.
An algorithm is presented for the problem of the stereopsis of time-varying images (the dynamic stereo problem). Dynamic stereopsis is the integration of two problems; static stereopsis and temporal correspondence. Rather than finding the intersectino of these problems to be more difficult, it was found that by solving the two problems simultaneously, and thus incorporating the spatio-temporal context within which a scene exists, some of the hard subproblems belonging to stereopsis and temporal correspondence could be avoided. The algorithm relies on a general smoothness assumption to assign both disparity and temporal matches. A simple model of the motion of three-dimensional features is used to guide the matching process and to identify conditional matches which vioate a general smoothness assumption. A spatial proximity rule is used to further restrict possible matches. The algorithm has been tested on both synthetic and real input sequences. Input sequences were chosen from three-dimensional moving light displays and from "real" grey-level digitized images.
Jenkin, M. and Kolers, P. A., Some problems with correspondence, RBCV-TR-86-10, Techinical Report on Research in Biological and Computational Vision, Department of Computer Science, University of Toronto, 1986
The notion of correspondence underlies many current theories of humand and machine visual information processing. Algorithms for both the correspondence process and solution to the correspondence problem have appeared regularly in the computer vision literature. Algorithms for stereopsis (Barnard & Thompson, 1980; Marr & Poggio, 1977; Mayhew & Frisby 1980) and for tracking objects through time (Dreschler & Nagel, 1981; Jain & Sethi, 1984; Moravec, 1977; Ullman, 1979; Webb, 1981) have been presented which assume that token matching of separated or successive views is the underlying visual process. This paper will address this notion of token matching as a primitive operation in vision. We will argue that correspondence seems ill suited to the task of accounting for how an object is poitioned in time or space, and that some other mechanism may provide a more apt account.
