The
ubiquitous video/audio project explores potential implementations of video and
audio technology that seeks to become transparent to everyday activities. As a
part of this project, we are developing video and audio processing techniques
and exploring their applications within the context of an “Aware Home.”
During the first year of the project, we developed prototypical systems for high
and low bandwidth video and audio processing.
In the second year we propose to explore
and prototype a spectrum of applications in the home that uses these systems.
We expect the use of video in the home to grow from isolated applications such as videophones, smart front doors, and surveillance devices to become pervasive. Such pervasive use of video and audio technology will make the entire volume of the house viewable, and permit synthesis of desired views from existing views. Once we have established an infrastructure to support such a pervasive use of video and audio technology, a homeowner will be able to install video hardware for one purpose and use it for many others. For this reason, we expect an explosion of demand for video applications. This proposal attempts to anticipate that demand and explore the technologies to provide ubiquitous sensing of video and audio and explore candidate video applications now. Making video and audio sensing transparent to everyday activities and distributing it everywhere in an environment is essential in the development of an Aware Home.
There are three
different classes of video and audio processing technologies that we are
exploring:
Low
cost, low bandwidth technologies that emphasize the transmission of single
images or low frame rates. These applications could use power lines or
wireless links to transmit images. Camera does little processing besides
reformatting data for transmission.
High frame rate applications that require the processing and transmission of images at 30 Hz or better. This is specifically important in applications where high fidelity information is required to track activities.
Camera
and sensors with local onboard processing.
These applications use task specific image processing right at the
camera. This increases the cost of each installed camera, but decreases the
requirement for high bandwidth information transmission.
In addition to the above enabling technologies, we are studying following methods to develop aware space (ie. Aware Home).
Haro, A., I. Essa, and M. Flickner, “A Non-invasive Computer Vision System For Reliable Eye Tracking”, In Proceedings of ACM CHI 2000, (Late Breaking Short Paper). The Hague, Netherlands, April 2000 (To Appear).
Haro, A., I. Essa, and M. Flickner, “Detecting and Tracking Eyes by Using their Physiological Properties, Dynamics and Appearance”, In Proceedings of IEEE CVPR 2000, December 2000 (To Appear).
Essa, I. “Ubiquitous Sensing for Smart and Aware Environments”, Submitted to IEEE Personal Communications, Special Issue on Networking the Physical World, January 2000.
Copyright © 1997-2000 |
Last Updated April 6, 2000.