Out-of-core adaptive iso-surface extraction from binary volume data.

Volumetric datasets have already been used in multiple domains. Recent improvements in acquisition devices have boosted the size of available datasets. We present an out-of-core algorithm for iso-surface extraction from huge volumetric data. Our algorithm uses a divide and conquer approach that divides the volume and processes every meta-cell sequentially. We combine our approach with a dual surface extraction algorithm in order to build adaptive meshes. Our solution produces patches of adaptive meshes that can finally be combined to generate a manifold and closed surface. As our approach processes only a part of the volume in-core, with a minimum of redundancy, it can handle very big volumes by modifying the meta-cells size to fit to the in-core memory available. Moreover, our algorithm can be parallelized in order to boost processing times and increase its interactivity. We present examples of the application of our solution to huge segmented volumes.