Presenter: Zhe Zhang
Advisor(s): Dr. David F. McAllister
Author(s): Zhe Zhang
Graduate Program: Operations Research

Title: Accelerating the Linear Optimization Calculation in Image Rendering

Abstract: A new stereo image rendering algorithm is introduced which in most cases has better color results than existing methods. In this algorithm to compute the color at a given pixel in the anaglyph image requires solving a linear programing problem. We exploit computational properties of the Simplex algorithm to reduce computation time by 75 to 80 percent. After computing the color at one pixel, a depth-first search is performed around it to collect all the pixels with color close enough to it. We use a certain criteria of measuring closeness of colors such that it is guaranteed the colors of all the collected pixels can be computed using a simple matrix-vector multiplication. We also parallelize the algorithm and implement it on a cluster environment. Different data dividing schemas are discussed and a dynamical load balancing technique is applied to reduce the total running time.