Browse Prior Art Database

Using video editing techniques to guarantee smooth playback of video summaries Disclosure Number: IPCOM000020341D
Publication Date: 2003-Nov-14
Document File: 1 page(s) / 28K

Publishing Venue

The Prior Art Database



This text was extracted from a Microsoft Word document.
This is the abbreviated version, containing approximately 74% of the total text.


Using video editing techniques to guarantee smooth playback of video summaries

         Work is ongoing to automatically generate video summaries by selecting key parts of the content and generating new play lists. The content of the video summaries is chosen by content analysis algorithms to ensure the summary has meaning to the user. In general when you choose arbitrary parts of an A/V sequence (e.g. coded as MPEG-2) and try to play them in sequence, there will be two problems: the MPEG content may not decode smoothly and the amount of data on disc may be too small to ensure continuous supply of data to the decoder and hence smooth video playback. The A/V content may also be stored in a fragmented way so even if each segment of the summary is of sufficient duration there is still a problem.

         When generating the video summary, the system should generate a new segment of A/V content (called a Bridge Sequence [1, 2]) and store it on disc. Within the Bridge Sequence, some A/V content can be re-encoded to ensure a smooth transition (in practice less than two MPEG-2 GOPs). Furthermore, extra A/V content can be copied to the Bridge Sequence to ensure that there is sufficient contiguous data on disc to ensure continuous playback. Another advantage of using Bridge Sequences is that it allows modification of the content at the edit point, for example, to include transitions (wipes, cross fades) or captions. This allows the generated summaries to be customized so it is clear they were...