Software Framework for Managing Heterogeneity in Mobile Collaborative Systems
Computer Supported Cooperative Work
| Hi-index | 0.00 |
Interactive applications on mobile devices often reducedata fidelity to adapt to resource constraints and variableuser preferences. In virtual reality applications, theproblem of reducing scene graph fidelity can be stated asa combinatorial optimization problem, where a part of thescene graph with maximum fidelity is chosen such that theresources it requires are below a given threshold and thehierarchical relationships are maintained. The problemcan be formulated as a variation of the Tree KnapsackProblem, which is known to be NP-hard. For this reason,solutions to this problem result in a tradeoff that affectsuser navigation. On one hand, exact solutions provide thehighest fidelity but may take long time to compute. On theother hand, greedy solutions are fast but lack highfidelity. We present a simplification architecture thatallows the exploration of such navigation tradeoffs. Thisis achieved by a formulating the problem in a generic wayand developing software components that allow thedynamic selection of algorithms and constraints. Theexperimental results show that the architecture is flexibleand supports dynamic reconfiguration.