Octree's breakthrough technology extends 3-D graphics into an additional spatial dimension, dramatically increasing performance and functionality while reducing costs. Now you can interact with HUGE 3-D data sets in.your application, presentations and webpages. Big Data Large sets of detailed 3-D information are becoming increasingly available from a variety of sources, such as terrestrial and airborne 3-D laser scanners, structured-light scanners, GPR (Ground-Penetrating Radar), medical scanners (CT, MR, etc.) as well as more traditional sources such as animation and mechanical-design systems. Effective utilization of this information can bring major benefits in many application areas, including:

• AEC (Architecture, Engineering & Construction)

• Medicine

• GIS (Geographical Information Systems)

• 3-D Biometrics (face matching, etc.)

• CAD/CAM (Computer-Aided Design & Manufacturing)

• Scientific visualization

• NDT (Non-Destructive Testing)

• Archaeology (heritage preservation, etc.)

• Inspection

• Mineral exploration

• Environmental

• Simulation (military, etc.)

Unfortunately, the technology used in the current generation of 3-D graphics systems was developed primarily for “toy” situations (video games) and becomes increasingly inefficient as data set size increases. Even with today’s 100,000,000+ transistor graphics chips the number of calculations required is a serious limitation.

Adding a Dimension for a New Generation of 3-D Graphics
Fortunately, this is not an inherent problem with large 3-D data sets, but only with these methods. Octree’s 3-D graphics technology was developed to solve this problem by deriving a new data representation that models an additional dimension. Models are represented as solids rather than surfaces. By equalizing the dimensionality of the data with the embedding space, models can be efficiently partitioned. This means we can now apply to 3-D the efficiency of the methods used so effectively in conventional databases for handling airline reservations, credit-card transactions, and so on. The result is a reduction in computations that becomes nothing less than astronomical as data sets become large, resulting in dramatic increases in performance and reductions in cost.

In addition, by directly modeling solids, we realize a quantum increase in functionality. For example, every model has an easily-computed volume, facilitating analysis, physics-based simulations, etc. Set operations become inexpensive, making clash and collision detection fast and reliable. Powerful volumetric operations (the 3-D equivalent of image processing) can be applied to sampled data.