Modifying data which has been coded

Coded data generation or conversion – Digital code to digital code converters – Adaptive coding

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C341S059000, C348S452000

Reexamination Certificate

active

06329934

ABSTRACT:

FIELD OF THE INVENTION
The invention relates to modifying data which has been coded. The invention may be applied, for example, to edit a sequence of pictures which has been coded in accordance with a standard defined by the Moving Pictures Experts Group (MPEG).
BACKGROUND ART
The document ISO/IEC 13818-2 relates to the MPEG-2 standard. Section 7 describes decoding steps for deriving a sequence of pictures from a data stream. These decoding steps include, amongst others, variable-length decoding, inverse quantization, inverse discrete-cosine transformation, and motion compensation. A sequence of pictures is coded in accordance with the MPEG-2 standard if a data stream is obtained which, by applying the decoding steps described in section 7, results in a correctly decoded sequence of pictures.
SUMMARY OF THE INVENTION
It is an object of the invention to modify data which has been coded in a cost-efficient manner. The invention takes the following aspects into consideration. In principle, it is possible to fully decode the data which has been coded, to modify the data, and to code the modified data. However, in some applications, such an approach will be relatively costly. This is the case, for example, if the data to be modified is a sequence of pictures which has been coded in accordance with a video-coding standard defined by the Moving Picture Experts GCroup (MPEG). MPEG coding or MPEG decoding of video data requires, amongst other things, a relative large data storage capacity for storing pictures. What is more, a full MPEG decoding followed by a full MPEG coding, involves a further motion estimation and compensation in addition to the motion estimation and compensation which have been carried out at the coding end. This may introduce distortion.
In accordance with the invention, data which has been coded is modified in the following manner. The data which has been coded is decoded only partially. That is, only a portion of the steps which are needed to fully decode the data, is carried out. Accordingly, partially decoded data is obtained. The partially decoded data is modified. The partially decoded data which has been modified, is coded so as to obtain coded modified data. It should be noted that because, in the invention, a modification is carried out on partially decoded data, rather than on fully decoded data, some distortion may be introduced. However, in many applications, the distortion introduced will be relatively small such that, nevertheless, a satisfactory quality can be obtained at a decoding end. Since, in the invention, only a partial decoding is carried out, fewer circuitry will be required than if the data to be modified is fully decoded. Thus, the invention allows data which has been coded, to be modified in a cost-efficient manner.
It has already been mentioned that the invention may be applied to edit a sequence of pictures which has been coded in accordance with an MPEG standard. In such an application, the partial decoding may comprise variable length decoding, inverse quantization and inverse discrete cosine transformation, but not motion compensation which requires a relatively large memory. As a result, blocks of prediction-error pixels will be obtained instead of blocks of picture pixels. The blocks of prediction-error pixels may be modified, for example, by adding blocks of auxiliary pixels to the blocks of prediction-error pixels, or by scaling the prediction-error pixels, or by filtering the blocks of prediction-error pixcels, or by any combination the aforementioned operations. The coding of modified blocks of prediction-error pixels comprises discrete cosine transformation, quantization and variable-length coding, but not motion estimation which is a complex operation. Thus, motion vectors re-used. Optionally, the coding may further comprise an error compensation for reducing any errors which may be introduced by the discrete cosine transformation and the quantization of a modified block of prediction-error pixels.
The invention and additional features which may be optionally used to implement the invention to advantage, are apparent from and elucidated with reference to the drawings described hereinafter.


REFERENCES:
patent: 4882583 (1989-11-01), Dimitri et al.
patent: 5481310 (1996-01-01), Hibi
patent: 5557276 (1996-09-01), Sakazawa et al.
patent: 5890125 (1999-03-01), Davis et al.
patent: 6181743 (2001-01-01), Bailleul
patent: 6246438 (2001-06-01), Nishikawa et al.
patent: WO-97/47128 (1997-12-01), None
patent: 2316568 (1998-02-01), None
Serial No. 09/275,362.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Modifying data which has been coded does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Modifying data which has been coded, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Modifying data which has been coded will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2593690

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.