ViCoMo |
System Architecture and Networking (SAN) |
Eindhoven University of Technology |
Introduction
The Video Context Modeling project addresses video content analysis, enhanced by modeling and using context information, and applying this in an embedded context. One part of the system concerns packing functionality into components that can be controlled through the network. Standard interfaces and methods to predict component performance must be developed in order to make predictable distributed applications. The work continues work in project Cantata in which we have looked at software structures that support tunable applications, and at platforms that manage resource budgets.
People Involved
SAN Contribution
TU/e SAN will supply the software infrastructure. This consists of two components:
Server software containing a distribution manager and application manager that enable a user to compose and deploy his surveillance application.
A middleware layer to be installed on top of the (real-time) operating system running on individual processors that constitute the hard-ware platform on which the surveillance application is running. Key component of this platform is a device manager that is capable to install and configure components in such a way that their service interface becomes exposed on the network, thereby becoming available for composition into a service oriented architecture. Another feature of this layer is the ability to claim the necessary resources from the hardware platform it runs on.
. ViCoMo component platform
Links
System Architecture and Networking | ||
Eindhoven University of Technology | ||
Last update: 19 Mar. 2010