PREVIOUS -- HOME  -- NEXT

3. Executive Summary:

Reason for Project

Recently, there has been greatly increased interest in integrating multimedia content into computer applications. One of the most challenging aspects of that task has been the effective use of video. The accurate representation of video information takes a tremendous amount of data. This amount of data is so great that it must be compressed to be effectively utilized by current computer hardware.

One particularly effective compression scheme is the international standard Motion Pictures Expert Group (MPEG) suite of algorithms. Although MPEG encoding allows for the compression of both audio and video over a range of qualities, this project concentrates on MPEG-1 video compression because it is specifically aimed at producing data rates optimized for computer communication and display.

Project Solution

One of the important properties of MPEG compression technique is that it is highly parallel. The approach of this project is to exploit that characteristic by performing the compression with software running on a group of inexpensive computers working together on a single video input.

The hardware configuration consists of three, networked, personal computers and a single monitor. Basically, the software creates a "virtual encoding machine" that operates at 60-80% of the sum of the speed of the individual machines. The software is controlled through a Unix command-line interface. The command line allows for easier automation of encoding, such as batch jobs. The resulting output is near real-time MPEG-1 video compression.

Project Results

Initially, there were three easily identifiable milestones selected to judge the success of this project.

  1. Creation of software that runs on a pre-configured network
  2. Manipulation of software to run on the specific hardware intended for the project
  3. Augmentation of software to work with the hardware to provide the fastest possible encoding process (real-time)
The project, originally scheduled to finish in December 1999, changed course forcing the pace to be intensified in order to reach completion by the end of the summer. This adjustment was done to accommodate other interests, namely early graduation for one of the team members. Additionally, this deadline change suggested a challenge similar to that found in industry. Modifications were made to the project to successfully complete it in partial accordance with the milestones.