Coding and decoding video signals

Television – Bandwidth reduction system – Data rate reduction

Patent

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

348414, H04N 728

Patent

active

058414703

DESCRIPTION:

BRIEF SUMMARY
FIELD OF THE INVENTION

This invention relates to a method of video coding in which video signals which comprise a series of frames supplied at an original frame rate are encoded by a first encoder at a first frame rate less than the original frame rate.
One application of this method of encoding video signals is in the field of videotelephony/videoconferencing over the public switched telephone network (PSTN).


BACKGROUND OF THE INVENTION

Current visual teleconferencing systems are based on the digital coding and transmission of television pictures. The user of a videotelephone is viewed by a video camera during the progress of a call, which camera outputs video signals at an original frame rate, for example in the region of 25 to 50 frames per second, to a video encoder.
In order to use low capacity digital networks for this purpose, significant compression of the image data is required. Study Group XV of the CCITT has recommended a coding scheme for low bit rate transmission of visual telephony signals at multiples of 64 kbits/s known as Recommendation H.261. This is a waveform coder and uses hybrid motion compensated interframe/DCT. H.261-based video codecs operating at 64 kbits/s are liable to produce low quality images; the frame rate is usually low and perceptually noticeable degradations occur when motion is large.
In order to avoid flicker in such low frame rate decoded moving images it is necessary to bring the display frame rate up to the order of 30 frames per second. This can be done by interpolating between the decoded frames or, more simply, by repeating a given decoded frame a number of times before refreshing the display with the next decoded frame. This technique is acceptable if the frames include moving images having little fast moving detail or fast moving detail which is of no interest to the viewer but in other cases the slow frame update can be a serious drawback.
An example of this in videotelephony applications arises with the movement of the mouth of the person speaking into the videotelephone. The mouth can make rapid excursions during speech the detail of which is lost at a coding rate of ten frames per second artificially increased in frame rate by successively repeating frames. This results in a loss of synchronisation between the lips and the speech greatly reducing the intelligibility of the transmission as the number of visual clues usually obtained from the mouth are greatly reduced.


SUMMARY OF THE INVENTION

According to a first aspect of the present invention a method of encoding video signals which comprise a series of frames supplied at an original frame rate in which the frames are encoded by a first encoder to provide a first coded output at a first frame rate less than the original frame rate is characterised in that one or more portions of the image area of the series of frames is/are encoded by a second encoder to provide a second coded output at a second frame rate greater than the first frame rate.
The first and second coded outputs can be decoded by a decoder according to a second aspect of the present invention and the resulting series of decoded images superimposed to provide a series of frames which is updated at the first frame rate with one or more selected portions updated at the second frame rate.
The portions of the series of frames may be those portions which correspond to one or more predetermined classes of image feature, for example the mouth or eyes of a human face.
Preferably, the second encoder encodes each of the portions of the series of frames by the method of principal components, the second coded output comprising a series of representative coefficients obtained from a portion of a frame corresponding to a predetermined class of image feature to be encoded and a respective reference set of vectors (eigenpictures). This provides compact coding of the portions of the frames as the principal components are of a well defined feature. If additional classes of image are also encoded by this method, each will be coded according to a respective

REFERENCES:
patent: 4488175 (1984-12-01), Netravali
patent: 4841575 (1989-06-01), Welsh
patent: 5086480 (1992-02-01), Sexton
patent: 5550580 (1996-08-01), Zhou
Aizawa et al, "A Model-Based Analysis/Synthesis Image Coding Scheme", Electronics and Communications in Japan, Apr. 1, 1990, No. 4, Part I, New York, US, pp. 200-207.

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

Coding and decoding video signals does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Coding and decoding video signals, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Coding and decoding video signals will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-1708732

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