This Small Business Innovation Research (SBIR) Phase I project seeks to develop a comprehensive visualization toolkit for massive 3D photography datasets. The objective is to create software that will become the standard of 3D photography for photorealistic modeling and visualization of large-scale scenes. Though the benefits of computer modeling are well-known, the bottleneck in this process has always been the difficulty in achieving rapid, seamless, data gathering and fusion to the model. Recent developments in range sensing have made possible the acquisition of accurate 3D scans of large scenes. The complexity of the acquired data sets, the large volumes of data collected, and the integration of 3D range data with conventional 2D color images is what drives this proposal. The goal is to minimize the effort of building models of high geometric and photometric accuracy that are suitable for efficient rendering, manipulation, and analysis. The core novelty to our research is the automated 3D-to-3D range-to-range registration, 2D-to-3D image-to-range registration, and integration of multi-view geometry to automated 3D registration for texture mapping unconstrained 2D images onto 3D range data. The system is to segment the range and color images, extract their features, establish topological relationships between features, and derive global topological relationships between features in the final 3D geometric model. The output will be a complete photorealistic model of the scene.

The development of a complete visualization toolkit based on the available registration and texture mapping modules and the newly proposed multi-view geometry module is the final product. Its chief advantage over existing approaches is that it merges the benefits of multi-view geometry with automated registration of 3D range scans to produce photorealistic models with minimal human interaction. This visualization work has obvious applications in any area where voluminous range scans and photographs must be acquired to produce complex 3D models. Increased use of 3D photography will have spillover benefits, including growing use of highly detailed 3D models in many other industries such as real estate, virtual tourism, and computer games.

Project Start
Project End
Budget Start
2006-07-01
Budget End
2007-06-30
Support Year
Fiscal Year
2006
Total Cost
$150,000
Indirect Cost
Name
Brainstorm Technology LLC
Department
Type
DUNS #
City
New York
State
NY
Country
United States
Zip Code
10001