Predictive Compression for Real Time
Multimedia Communication Using Eye Movement Analysis
Oleg Komogortsev and
Javed I. Khan
Human Computer Interaction Laboratory
Department of Computer Science
Texas State University-San Marcos
Last Revised May 20, 2009
Approximately
2 degrees in our 180 degree vision span has sharp vision. Many researchers have
been fascinated by the idea of eye-tracking integrated perceptual compression
of a media stream, yet any practical system has yet to emerge. The unique
challenge presented by real time media compression is how to handle the fast
nature of the human eye movements and provide its integration with
computationally intensive transcoding scheme. The delay introduced by the media
transmission in the network presents a unique challenge for any practical
system. This paper presents an approach which solves these problems.
This report contains experiment clips used to test the performance of this system. The videos are MPEG-2 ISO 13818-2 streams. We recommend using MPlayer found at www.mplayerhq.hu or Winamp to view these video samples.
Video Samples:
|
Sample Name |
Original Sample |
Perceptual Attention Focus Window (WPAW) with eye-gaze displayed |
Perceptually Encoded |
|
Car |
|||
|
Shamu |
|||
|
Airplanes |