The O-buffer is basically a conventional image or volume except that the positions of sampling points are recorded by offsets to a regular grid. The regular grid can be rectangular, cylindrical, spherical, etc. The O-buffer can be either 2D or 3D. The number of samples in every grid cell of an O-buffer can be the same or different.
The O-buffer can be classified into three categories:
- Uniform O-buffer : The number of offset samples stored in every grid cell of the O-buffer is the same
- Nonuniform O-buffer : The number of offset samples stored in every grid cell of the O-buffer is not the same
- Adaptive O-buffer : The regular grid is adaptive.
This projects involves
- Investigate the basic forms, quantization, data structures, and construction of the O-buffer
- Develop the fundamental algorithms for the O-buffer such as rendering and multi-resolution
- Explore applications of the O-buffer in flythrough systems, network graphics, solid modeling, and medical VR system
- Develop a software system (ObVis) which will integrate our developments related to O-buffers
This project is supported by NSF grant CCR-0306438.