|
|
|
Introduction
|
Project Name :
|
Flexible Framework for GPGPU-based Video Decoding and Post-processing. (ITS/139/09)
|
|
|
|
|
Objective :
|
The objective of this project is to develop a flexible framework and toolset for utilizing the
General Purpose Graphics Processing Unit (GPGPU) in high-definition video decoding and
post-processing. Although there are software packages which make use of the GPGPU for
high-definition video decoding and post-processing, these software packages, however, are
usually proprietary and not flexible enough to allow further applications to be built upon them.
Besides, from technical perspective most of the existing solutions focused on employing
GPGPU to perform one single video stream decoding, without noting that there might be cases
where multiple video streams, for instance in video surveillance, have to be decoded
simultaneously. In this project, we will have a deeper exploration on how to utilize GPGPU in a
more flexible way, and to define a framework and to develop a toolset so that multiple
high-definition video streams decoding and post-processing can be properly offloaded to the
GPGPU. With this toolset, potential applications including high-definition video editing,
transcoding, etc. can be realized, but it can be ascertained that video surveillance industry in
Hong Kong can be directly benefited by the toolset developed in this project as it addresses the
fundamental problems in decoding multiple high-definition video streams on a typical
workstation.
|
|
|
Project Duration :
|
12 months
|
 |
visits |
|
|
|
|
Introduction