Prefetching Optimization for Distributed Urban Environments

We propose a method for pre-fetching scene data to improve smoothness of client side rendering in distributed urban environments. The scene is represented as a graph where the nodes are the view-cells and the edges denote the adjacencies among the cells. Prefetching is transformed into graph partitioning problem where the partitions might overlap with each other. We utilize simulated annealing algorithm to find an approximation for the optimum partitioning where the rendering cost is minimum. The presented method works on heuristics to minimize average cost and maximum view-cell transition cost.