23 August 2001 Evaluation and extension of SGI Vizserver
Author Affiliations +
Abstract
SGI Vizserver enables remote visualization in a manner transparent to the user application by producing rendered output at geographically remote locations while utilizing the powerful pipeline and expansive memory of an Onyx2 Infinite Reality machine located at some centralized place. Since the communication of visualization imagery typically requires enormous bandwidth, Vizserver offers two built-in options for compression which provide high-quality images at interactive frame rates for local-area networks with bandwidths of 100 Mbps. However, these built-in compressors are not well suited to truly remote users who are separated from the server by great distances and connected through low- and very-low-bandwidth links. In this paper, we propose two external compression algorithms that connect to Vizserver via an API to achieve 1) greater flexibility in terms of the user's control over distortion and bandwidth performance, and 2) better overall performance for truly remote users. Of all the techniques considered, we find that a simple frame- differencing scheme is best suited to very-low-bandwidth operation in that it achieves visually lossless performance at a frame rate higher than that of the built-in options, which saturate the network and incur substantial amounts of frame dropping.
© (2001) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Tong Chu, Tong Chu, James E. Fowler, James E. Fowler, Robert J. Moorhead, Robert J. Moorhead, } "Evaluation and extension of SGI Vizserver", Proc. SPIE 4368, Visualization of Temporal and Spatial Data for Civilian and Defense Applications, (23 August 2001); doi: 10.1117/12.438110; https://doi.org/10.1117/12.438110
PROCEEDINGS
11 PAGES


SHARE
RELATED CONTENT

Perceptual wavelet scheme for image compression
Proceedings of SPIE (June 07 2001)
Spatial quantization via local texture masking
Proceedings of SPIE (March 17 2005)
New feature-based image adaptive vector-quantization coder
Proceedings of SPIE (December 07 1995)
Study of edge-based lossy image compression algorithms
Proceedings of SPIE (August 21 1995)

Back to Top