Encoding of two correlated sequences of data

Details Encoding of two correlated sequences of data Encoding of two correlated sequences of data

2001-03-30

2004-12-14

Philippe, Gims (Department: 2613)

Pulse or digital communications

Bandwidth reduction or expansion

Television or motion video signal

C382S238000, C348S409100

Reexamination Certificate

active

06831950

ABSTRACT:

FIELD OF THE INVENTION
The invention relates to a method of encoding a first sequence of data blocks and a second sequence of data blocks that is correlated with the first sequence of data blocks. The invention may be applied, for example, to encode stereoscopic (3D) video.
BACKGROUND OF THE INVENTION
It is possible to encode a sequence of data blocks in such a manner that certain data blocks serve as a reference for predictive encoding of the other data blocks. This technique is applied, for example, for coding video frames in accordance with a standard of the Moving Picture Expert Group (MPEG).
There are applications in which two correlated sequences of data blocks need to be coded. The coding of stereoscopic video is an example. Stereoscopic (3D) video will generally comprise a sequence of left-eye video frames and a sequence of right-eye video frames. The sequence of left-eye video frames and the sequence of right-eye video frames can be seen as a first and a second sequence of data blocks, respectively. The sequences are correlated.
Stereoscopic video can be encoded in the following manner. The sequence of left-eye video frames is encoded in accordance with an MPEG video coding technique as if this sequence were an ordinary video signal. This implies that certain left-eye video frames will serve as a reference for predictive coding of the other left-eye video frames. As for the sequence of right-eye video frames, each right-eye video frame is predictively encoded with respect to the corresponding left-eye video frame. This means that each left-eye video frame forms a reference for predictive encoding of the corresponding right-eye video frame. This manner of encoding stereoscopic video seems to be disclosed in the European patent application published under number 0 639 031.
SUMMARY OF THE INVENTION
It is an object of the invention to allow a cost reduction to be achieved.
The invention takes the following aspects into consideration. In order to decode a data block that has been predictively encoded, it is necessary that each data block that has served as a reference for this predictive encoding is available. Consequently, at a decoding end, it will be necessary to temporarily store data blocks that have served as a reference for the predictive encoding of other data blocks.
In accordance with the prior art, each right-eye video frame is predictively encoded with respect to the corresponding left-eye video frame. This implies that each left-eye video frame needs to be temporarily stored at the decoding end. This storage requires a relatively large memory, in particular because video frames generally comprise relatively many bits.
In accordance with the invention, a first sequence of data blocks is encoded in such a manner that certain data blocks serve as a reference for predictive encoding of the other data blocks. A second sequence of data blocks, which is correlated with the first sequence of data blocks, is encoded in such a manner that all data blocks are predictively coded with respect to the data blocks in the first sequence that serve as a reference.
Accordingly, at the decoding end, it will not be necessary to store all data blocks comprised in the first sequence whereas, in contrast, this is necessary in the prior art. It is sufficient to store those data blocks in the first sequence that served as a reference. Consequently, a decoder in accordance with the invention will require less storage capacity than a decoder in accordance with the prior art. Consequently, the invention allows a cost reduction to be achieved.
Another advantage of the invention relates to the following aspects. Predictive encoding inherently entails some inaccuracies: a prediction is rarely 100% correct. These inaccuracies will generally have an adverse effect on the quality of the data that will be obtained at a decoding end. In accordance with the prior art, some data blocks in the second sequence are predictively encoded with respect to data blocks in the first sequence that are themselves results of predictive encoding. This means that, in accordance with the prior art, the encoding of some data blocks involves two predictive encoding operations in series. Consequently, inaccuracies will accumulate. Since, in accordance with the invention, all data blocks in the second sequence are predictively coded with respect to the data blocks in the first sequence that serve as a reference, such an accumulation of inaccuracies cannot occur. Consequently, the invention enables a better quality to be obtained.
These and other aspects of the invention will be described in greater detail hereinafter with reference to the drawings.


REFERENCES:
patent: 5602593 (1997-02-01), Katto
patent: 5748248 (1998-05-01), Parke
patent: 5778099 (1998-07-01), Suzuki
patent: 5982439 (1999-11-01), Parke
patent: 6466620 (2002-10-01), Lee
patent: 06390381 (1995-02-01), None

Affiliated with

Also associated with

Rating

Encoding of two correlated sequences of data does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Encoding of two correlated sequences of data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Encoding of two correlated sequences of data will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3287687