This project will combine sets of lighting basis images gathered automatically with sparse 3D data gathered manually to form a "thick" photo. "Thick" photography methods are seen as a fast, simple, and low cost adjunct to existing conventional photography. Merging image-based-relighting (IBR) and computer vision techniques, the project plans to create new these new photographic images by (1) illuminating the object with a novel computer controlled light source as we gather basis images automatically (2) measuring a small set of widely spaced points on the object by touching them with a calibrated laparoscopic camera attached to a COTS 3D stylus digitizer, (3) photographing interior and occluded views with the laparoscopic camera. As envisioned, users can then: (1) interactively re-light a "thick" photo to explore surface textures, and clarifying shape and occlusions; (2) interactively change the camera viewpoint by modest amounts as the lighting remains stationary; and (3) interactively reveal important but occluded object features, such as the internal structures of Moche' pottery, the subject of this experimental effort..The research and scholarly impact can be significant. Many current image digitizing projects are now underway, but gather only softly lit, high quality digital photographs. These are of limited use as objects of study because they offer no simple way to explore the objects in a way that clarifies its shape, material, or internal structure. While texture-mapped 3D scanning methods enable users to explore and compare hard-surfaced objects, scanning is far more complex, slow and expensive than photography and visual quality is uneven. The value of digital archives of museum and library artifacts can increased greatly as a result of the proposed research. With museum partners, the researchers have selected the pottery of the Moche culture in Peru as the image testbed; these ceramics are found in museums across the United States and the world. Their interior structures are culturally significant but occluded, and their external shapes are simple but with a dark and highly reflective finish they can form a particularly difficult challenge for conventional 3-D scanning or photography. The project will employ devices for data gathering that is low-cost, accessible and simple, thus lowering the barriers for deployment at any museum.

Agency
National Science Foundation (NSF)
Institute
Division of Information and Intelligent Systems (IIS)
Type
Standard Grant (Standard)
Application #
0535236
Program Officer
Sylvia J. Spengler
Project Start
Project End
Budget Start
2005-12-01
Budget End
2010-11-30
Support Year
Fiscal Year
2005
Total Cost
$328,874
Indirect Cost
Name
Northwestern University at Chicago
Department
Type
DUNS #
City
Evanston
State
IL
Country
United States
Zip Code
60201