Method and apparatus for processing compressed video data streams

Abstract

The present invention relates to the field of digital broadcasting, and more particularly the insertion of digital video streams into other digital video streams. Compressed digital video streams, such as those compressed using the common MPEG-2 system, use a sequence of frames to compress a video sequence. Part of the encoding method to compress frames involves making predictions based on past or future frames. Where part of a compressed video stream is to be inserted into another existing video stream, problems may arise at the insertion point due to dependencies on past or future frames which occur outside of the insertion point. The effect of this is that the decoding process lacks information on which to make its predictions, and this could cause a decoder to reset or display frames out of order. The present invention overcomes this problem in a way which allows frame accurate insertion to be achieved without compromising quality. The present invention can be used to pre-process a compressed video stream ready for insertion, or can be used to dynamically insert a compressed video stream into an existing compressed video stream.