what is compressed domain video editing?

[sybase2008]

does it means after the editing, the video will not be re-encoding?

thanks

[guns1inger]

The idea is to minimise the need to re-encode by working with the video data in it's compressed state. Some NLEs - Vegas, Premiere for example - must decompress the video data to be able to work with it. Other editors, usually dedicated editors for specific types of video compression, work with the video in it's compressed state.

However, because some video compression (certainly mpeg1, 2 and 4) uses temporal compression extensively, it is difficult, if not impossible, to do all editing without some re-encoding. Certainly, any edit between the start and end of a group of pictures will require some degree of re-encoding to rebuild the data that would have been previously interpreted from now missing frames.

Also, any new data being introduced to the video stream will require some degree of re-encoding. At the present time this includes pretty much anything that alters the existing data stream, from simple adjustments through to major changes.

If you google Compressed Domain Editing you will find a lot of research on methods to improve working in this type of environment, as it is will continue to be the most prevalent form of compression for video for some time to come.

[JohnnyMalaria]

Here's an excellent paper that describes compressed domain processing:

http://www.hpl.hp.com/techreports/2002/HPL-2002-282.pdf