Performing multicut on walkable environments: Obtaining a minimally connected multi-layered environment from a walkable environment
Publication date
2016
Editors
Li, Minming
Wang, Lusheng
Chan, T-H. Hubert
Advisors
Supervisors
Document Type
Part of book
Metadata
Show full item recordCollections
License
Abstract
A multi-layered environment is a representation of the walkable space in a 3D virtual environment that comprises a set of twodimensional layers together with the locations where the different layers touch, which are called connections. This representation can be used for crowd simulations, e.g. to determine evacuation times in complex buildings. Since the execution times of many algorithms depend on the number of connections, we will study multi-layered environments with a minimal number of connections. We show how finding a minimally connected multi-layered environment can be formulated as an instance of the multicut problem. We will prove that finding a minimally connected multi-layered environment is an NP-Hard problem. Lastly, we will present techniques which shrink the size of the underlying graph by removing redundant information. These techniques decrease the input size for algorithms that use this representation for finding multi-layered environments.
Keywords
Simple Path, Visibility Graph, Graph Reduction, Crowd Simulation, Walkable Environment, Taverne, Theoretical Computer Science, General Computer Science
Citation
Hillebrand, A, van den Akker, M, Geraerts, R & Hoogeveen, H 2016, Performing multicut on walkable environments : Obtaining a minimally connected multi-layered environment from a walkable environment. in M Li, L Wang & T-H H Chan (eds), Combinatorial Optimization and Applications - 10th International Conference, COCOA 2016, Proceedings. vol. 10043, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10043 LNCS, Springer, pp. 311-325, 10th Annual International Conference on Combinatorial Optimization and Applications, COCOA 2016, Hong Kong, China, 16/12/16. https://doi.org/10.1007/978-3-319-48749-6_23, https://doi.org/10.1007/978-3-319-48749-6_23, conference