Asish Law

Hardware assisted volume rendering of unstructured grids by incremental slicing

Some of the more important research results in computational science rely on the use of simulation methods that operate on unstructured grids. However, these grids, composed of a set of polyhedra, introduce exceptional problems with respect to data visualization. Volume rendering techniques, originally developed to handle rectangular grids, show significant promise for general use with unstructured grids as well. The main disadvantage of this approach, compared to isosurfaces, particles or other visualization tools is its non-interactive performance. We describe an efficient method for rendering unstructured grids that is based on incremental slicing and hardware polygon rendering. For a given view direction, the grid vertices are transformed to image space using available graphics hardware. We then incrementally compute the 2D polygon-meshes that result from letting a set of planes, parallel to the screen plane, intersect (slice) the transformed grid. Finally, we use the graphics hardware to render (interpolate-fill) the polygon-meshes and composite them in visibility order. We show that, in addition to being faster than existing methods, our approach also provides adaptive control and progressive image generation. The adaptive method provides user control to ensure that the contribution of every cell is included in the final image or to limit the number of cells that are missed.

An Optimal Ray Traversal Scheme for Visualizing Colossal Medical Volumes

Modern computers are unable to store in main memory the complete data of high resolution medical images. Even on secondary memory (disk), such large datasets are sometimes stored in a compressed form. At rendering time, parts of the volume are requested by the rendering algorithm and are loaded from disk. If one is not careful, the same regions may be (decompressed and) loaded to memory several times. Instead, a coherent algorithm should be designed that minimizes this thrashing and optimizes the time and effort spent to (uncompress and) load the volume. We present an algorithm that divides the volume into cubic cells, each (compressed and) stored on disk, in contrast to the more common slice-based storage. At rendering time, each cell is allocated a queue of rays. For a sequence of images, all rays are spawned and queued at the cells they intersect first. Cells are loaded, one at a time, in front-to-back (FTB) order. A loaded cell is rendered by all rays found in its queue. We analyze the algorithm in detail and demonstrate its advantages over existing ray casting volume rendering methods.

The active-ray approach to rendering on distributed memory multiprocessors

Object dataflow is a popular approach used in parallel rendering. The data representing the 3D scene is statically distributed among processors and objects are fetched and cached only on demand. Most previous object dataflow methods were implemented on shared memory architectures and exploited spatial coherency to reduce hardware cache misses. We propose an efficient model for object dataflow parallel volume rendering on message passing machines. The active ray tracing algorithm is introduced and its ray storage mechanism is used to support latency hiding by postponing computation on inactive rays. Memory usage is optimized by letting objects migrate and replicate at different processors rather than the common static assignments. Our cache-only-memory approach uses a distributed-directory scheme to trace the location of objects at other nodes. A mechanism to minimize network congestion was implemented which optimizes channel utilization. Unlike previous methods, our approach can benefit from temporal coherence and effectively minimizes communication costs in successive frames. We implemented a volume ray casting instance of the algorithm on the Cray T3D and achieved higher efficiency and scalability than existing algorithms. We achieve interactive frame rates of approximately 20 Hz for 128/sup 3/ volume, and 4 Hz for 256/sup 3/ volume on 128 processors.

A memory-optimized visualization system for limited-bandwidth multiprocessing environments

Object dataflow is a popular approach used in parallel rendering. The data representing the 3D scene is statically distributed among processors and objects are fetched and cached only on demand. Most previous object dataflow methods were implemented on shared memory architectures and exploited spatial coherency to reduce hardware cache misses. We propose an efficient model for object dataflow parallel volume rendering on message passing machines. The algorithm is introduced and its ray storage mechanism is used to support latency hiding by postponing computation on inactive rays. Memory usage is optimized by letting objects migrate and replicate at different processors rather than the common static assignments. Our cache only memory approach uses a distributed directory scheme to trace the location of objects at other nodes. A mechanism to minimize network congestion was implemented which optimizes channel utilization. Unlike previous methods, our approach can benefit from temporal coherence and effectively minimizes communication costs during animation on limited bandwidth multiprocessing environments. We report results of the algorithms implementation on several platforms like Cray T3D, Convex SPP and DEC alpha cluster of workstations (COWs), and achieved higher efficiency and scalability than existing algorithms.