| Method for Efficient DCT Domain Mode Prediction for H.264 Video Transcoding 
Description The communication and entertainment industries have been profoundly changed by advances in digital technology. Much of the advancement can be directly attributed to techniques for handling video files using novel coding algorithms for video compression. Most of the video used in digital entertainment applications uses MPEG-2 video coding algorithm. MPEG-4, a more recent addition in the MPEG series, is a video format used for streaming over the Internet and used with mobile phones. The recently developed H.264 video is very efficient compressing video to one-half of the MPEG-2 bitrates at the same quality. The H.264 video standard is expected to replace MPEG-2 over the next several years and is also expected to appear in mobile devices soon because of the substantial bandwidth and quality advantages over MPEG-4. One issue that remains largely unresolved is efficient ways of converting H.263, MPEG-2, or MPEG-4 video to H.264 format. This invention addresses this issue by creating a fast and effective way of transcoding DCT based video such as MPEG-2, MPEG-4, and H.263 to the recently standardized H.264 video. The proposed approach exploits the DCT information gathered during the decoding stage to reduce the transcoding complexity. Potential Applications
Transcoders for MPEG-2, MPEG-4 or H.263 to H.264 format including: - Applications in developing high density transcoders for telecommunications applications such as video messaging.
- Video distribution from digital TV to the mobile device (in-home distribution or a remote delivery service).
- Transcoding in a DVR which saves space as H.264 takes fewer bits to compress the same video
State of Technology Development and Commercialization
Reference implementation is complete. Results show significant drop in computations complexity with a negligible drop in quality. 
(c) Florida Atlantic University All rights reserved.
Updated August 2008 | 
