This community resource development project provides trace characterizations of novel video encodings for performance evaluations of communication networks. For instance, the new H.264 video codec is dramatically changing multimedia networking through its increased coding efficiency (typically half the average bit rate compared to an MPEG-4 encoding with same video quality) and drastically increased bit rate variability. In this project, we design trace characterizations of the new video codecs, generate a large set of video traces, and distribute the resulting video trace library to networking researchers. Effective usage guidelines are critical for the video trace library to ensure that networking researchers, faculty and students alike, can quickly incorporate video traces in their networking studies. The usage guidelines are to a large degree based on statistical analyses of the generated video traces, providing an overview of the bit rate, video quality, and video frame time dependencies. The two pre-selected students, Stephen Charnicki and Jonathan Vahabzadeh, will be mainly involved in the statistical analysis and the preparation of usage guidelines for the video trace library. In preparation for their summer research, Stephen Charnicki and Jonathan Vahabzadeh have already attended a number of our group meetings and immersed themselves in the research literature on video traces. According to their personal interests they have already identified specialization areas that they will focus on, namely Stephen Charnicki will focus on High Definition (HD) video characterizations and Jonathan Vahabzadeh will focus on frame dependencies in the novel H.264 Scalable Video Coding (SVC). Within their respective focus areas, in consultation and collaboration with the project PIs and the graduate students working on the project, they will primarily be involved in the following research tasks:
|Effective start/end date||4/1/08 → 3/31/14|
- National Science Foundation (NSF): $677,999.00
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.