Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
2002-03-05
2003-12-23
Britton, Howard (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240250
Reexamination Certificate
active
06668018
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to methods and apparatus for decoding video data and, more particularly, to methods and apparatus for improving the quality of images generated by reduced resolution video decoders and to new and improved video decoders which produce reduced resolution images from encoded video data.
BACKGROUND OF THE INVENTION
The use of digital signals to transmit video data is becoming ever more common. In order to efficiently transmit video information, data compression is often used.
The International Standards Organization has set a standard referred to as MPEG that is intended to be used for the encoding and transmission of digital video signals such as high definition television (HDTV). One version of the MPEG standard, MPEG-2, is described in the International Standards Organization—Moving Picture Experts Group, Drafts of Recommendation H.262, ISO/IEC 1318-1 and 1318-2 of November 1994.
MPEG-2 and similar encoding systems rely on the use of discrete cosine transform coding and motion compensated prediction techniques, e.g., the use of motion vectors, to reduce the amount of digital data required to represent a series of digital images. Motion compensated prediction involves the use of video data transmitted to represent one frame, e.g., an anchor frame, to reconstruct subsequent, e.g., predicted frames, at decoding time.
Motion vectors are commonly used to implement motion compensated video coding. A motion vector can be used to represent a portion, e.g., a square or rectangular group of pixels, in a predicted frame. A motion vector includes information identifying a group of pixels in a reference frame to be used during video decoding in the generation of a current video frame. A motion vector includes vertical and horizontal offsets, interpreted relative to the position of a current macroblock of video data being decoded. The offsets included in a motion vector identify the pixel data in a reference frame to be used in generating the current frame. In MPEG 2, a macroblock corresponds to a group of 16×16 pixels. In the present application the term “macroblock” is used in a manner that is consistent with its MPEG-2 meaning. The term “block” or the phrase “block of pixels” is intended to refer to any group of pixels and is not intended to be limited to an MPEG-2 block which normally corresponds to a group of 4 pixels.
It is anticipated that at least some video images will be transmitted at resolutions far in excess of those commonly used today for NTSC television broadcasts. Television broadcasts at, e.g., resolutions of 1080×1920 pixels, are often referred to as high definition television signals because they exceed the resolution of current NTSC television images. Television broadcasts involving the transmission of images having a resolution that is the same as or similar to present NTSC television signals are commonly referred to as standard definition television (SDTV) broadcasts.
The amount of data which must be stored and processed for HDTV signals can be considerably greater than that for SDTV signals. Because of the amount of memory and the processing power required to decode HDTV signals in real time, HDTV decoders can be considerably more expensive than SDTV decoders.
The use of reduced resolution video decoders has been suggested in order to allow HDTV video signals to be decoded and displayed using video decoders and display devices which are generally comparable in cost to SDTV decoders. Reduced resolution video decoders, also sometimes referred to as downsampling video decoders, reduce the amount of data used to represent video images thereby also reducing the amount of memory and processing power required to decode an HDTV signal. The decoding of an HDTV signal using a reduced resolution decoder results in the generation of, e.g., SDTV resolution images from an encoded HDTV signal.
Referring now to
FIG. 1
, there is illustrated a video decoder which is representative of various known reduced resolution video decoders. The reduced resolution video decoder includes an (optional) preparser
112
, a syntax parser and variable length decoder circuit
120
, an inverse quantization circuit
122
, and an inverse DCT circuit
124
. The output of the inverse DCT circuit
124
is coupled to the input of a downsampler
126
. The downsampler
126
is used to reduce the resolution of the video images being processed and thus the amount of decoded video data which is stored in a video memory
114
, e.g., for use as reference frames when generating subsequent frames encoded using motion vectors. In addition to the reference frame memory
114
, the video decoder
100
includes a switch
129
, summer
128
, a pair of motion compensated prediction module
131
and a select/average predictions circuit
134
. The motion compensated prediction modules
131
perform uni-directional predictions. In order to form bi-directional predictions, e.g., when processing B-frames, both motion compensated prediction modules
131
are used with the output of the tow modules
131
being combined by the select/average predictions circuit. A single one of the prediction modules
131
is used when performing uni-directions predictions, e.g., when processing P-frames with the select/average predictions circuit
134
selecting the output of the appropriate module
134
to be used. The pixel data generated from reference frame data and output by the select/average predictions circuit
134
is combined by the summer
128
with received decoded video data to generate a complete representation of a video frame which was encoded using motion compensated prediction.
Because HDTV motion vectors are intended to be applied to full HDTV resolution video images and not downsampled video images such as those stored in the memory
114
, the motion compensated prediction modules
131
must perform reduced resolution prediction prior to the data being combined by the summer
128
, so that the data generated by the motion compensated prediction modules
131
will be of the same reduced resolution as the data output by the downsampler
126
.
For a detailed discussion of various reduced resolution video decoders capable of decoding HDTV digital video data see U.S. Pat. No. 5,614,952 which is hereby expressly incorporated by reference.
While reduced resolution video decoders have significant advantages over HDTV decoders in terms of cost, the images generated by such decoders can suffer not only in terms of a reduction in resolution corresponding to the amount of downsampling performed on the HDTV image but also in terms of picture degradation resulting from the use of motion vectors. The use of motion vectors by reduced resolution video decoders offers the potential for serious image degradation in some instances resulting from prediction errors. Such prediction errors are due in large part to the application of motion vectors which were encoded to be used with full resolution reference frames being applied to reduced resolution, e.g., downsampled reference frames.
Thus, the use of known downsampling decoders can lead to certain annoying picture artifacts under particular combinations of scene content and motion vector conditions.
In order to produce decoded video images having a high degree of quality using a reduced resolution decoder, there is a need for methods and apparatus for identifying scene conditions and motion vectors which may result in significant and/or annoying prediction errors and thus degrade image quality. In addition, there is a need for methods and apparatus which can eliminate or minimize the degree of picture degradation resulting from the processing or use of motion vectors by reduced resolution video decoders.
SUMMARY OF THE PRESENT INVENTION
The present invention relates to methods and apparatus for improving the quality of images generated by reduced resolution video decoders and to new and improved video decoders which produce reduced resolution images from encoded video data.
Various features and embodiments of the present invention a
Fuhrer Jack
Henderson John
Pearlstein Larry
Britton Howard
Straub Michael P.
Straub & Pokotylo
LandOfFree
Methods and apparatus for representing different portions of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for representing different portions of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for representing different portions of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3131613