Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1999-01-22
2002-05-21
Le, Vu (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240160, C375S240210
Reexamination Certificate
active
06393059
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to conversion of bit stream of video data in highly efficient coding. More specifically, this invention relates to conversion between bit streams of video data encoded by interframe predictive coding at different block sizes or at different levels of motion compensation accuracy.
There is a need to convert bit streams of moving pictures coded by efficient coding under MPEG (Moving Picture Experts Group) standard into bit streams of different data rates or to convert bit streams of variable data rates into bit streams of fixed data rates.
In such cases, video data (original picture) is generally completely decoded and encoded again to form video data of a different data rate. A portion of the video data to be converted can be used as it is for conversion when encoding processing is substantially the same between encoding of the original picture and re-encoding in conversion. In detail, motion vector (MV) data for the video data to be converted is used for re-encoding as it is, so that motion vector detection with complicated calculation for the re-encoding can be omitted.
Furthermore, motion-compensated interframe predictive processing is the same between encoding of the original picture and re-encoding in conversion. Degradation of picture quality due to re-encoding depends only on quantization for the re-encoding, thus minimizing the degradation of picture quality.
The above processing is described in the paper “A Study on Coding Control Scheme for Video Transcoding”, 1-6, published for 1993 Moving Picture Symposium (PCSJ93).
Conversion of moving picture bit streams is conventionally applicable only to bit streams of the same pixel block size that undergoes motion compensation between encoding of original picture and re-encoding in conversion. The same MV data can be used for encoding of original picture and re-encoding in conversion as described above. However, the same MV data cannot be used for the conversion when the block sizes that undergo motion compensation are different from each other between the encoding of the original picture and the re-encoding in conversion. Conversion at different pixel block sizes for motion compensation requires complete decoding and re-encoding. This causes an increase in the amount of data to be processed and a decrease in the quality of pictures after conversion.
Furthermore, conversion of moving picture bit streams is conventionally applicable to bit streams to be processed by the same encoding processing between encoding of original picture and re-encoding in conversion. In such cases, the same MV data can be used for the encoding of the original picture and the re-encoding. However, the same MV data cannot be used for conversion when the encoding processing or motion compensation accuracy are different from each other between the encoding of the original picture and the re-encoding in conversion. Conversion at different encoding processing requires complete decoding and re-encoding. This also causes an increase in the amount of data to be processed and a decrease in the quality of pictures after conversion.
SUMMARY OF THE INVENTION
A purpose of the present invention is to provide a moving picture bit stream conversion apparatus and a method thereof that realize no increase in the amount of data to be processed and no decrease in the quality of pictures after conversion, even when the block sizes that undergo motion compensation or encoding processing are different from each other between encoding of an original picture and re-encoding in conversion.
The present invention provides an apparatus for converting a first bit stream of moving picture that has been encoded by motion-compensated prediction to a second bit stream of moving picture. The apparatus includes a separator to separate a bit stream of first motion vector data from the first bit stream. The first motion vector data corresponds to a first block size. The apparatus also includes a decoder to decode a portion or all of the first bit stream by motion-compensated prediction at the first block size using the first motion vector data to obtain a decoded picture or a decoded signal at an intermediate processing stage. Also included by the apparatus is a forming unit to form a bit stream of second motion vector data using the first motion vector data. The second motion vector data corresponds to a second block size that is different from the first block size. The apparatus still includes an encoder to encode the decoded picture or the decoded signal by motion-compensated prediction at the second block size using the second motion vector data to obtain a third bit stream, and also a multiplexer to multiplex the third bit stream and the bit stream of the second motion vector data to obtain the second bit stream.
Furthermore, the present invention provides a method of converting a first bit stream of moving picture that has been encoded by motion-compensated prediction to a second bit stream of moving picture. A bit stream of first motion vector data is separated from the first bit stream. The first motion vector data corresponds to a first block size. A portion or all of the first bit stream is decoded by motion-compensated prediction at the first block size using the first motion vector data to obtain a decoded picture or a decoded signal at an intermediate processing stage. A bit stream of second motion vector data is formed using the first motion vector data. The second motion vector data corresponds to a second block size that is different from the first block size. The decoded picture or the decoded signal is encoded by motion-compensated prediction at the second block size using the second motion vector data to obtain a third bit stream. The third bit stream and the bit stream of the second motion vector data are multiplexed to obtain the second bit stream.
Furthermore, the present invention provides an apparatus for converting a first bit stream of moving picture that has been encoded by motion-compensated prediction to a second bit stream of moving picture. The apparatus includes a separator to separate a bit stream of first motion vector data from the first bit stream. The first motion vector corresponds to first accuracy of motion compensation. The apparatus also includes a decoder to decode a portion or all of the first bit stream by motion-compensated prediction at the first accuracy of motion compensation using the first motion vector data to obtain a decoded picture or a decoded signal at an intermediate processing stage. Further included by the apparatus is a forming unit to form a bit stream of second motion vector data by motion-compensated prediction using the first motion vector data. The second motion vector data corresponds to second accuracy of motion compensation that is different from the first accuracy of motion compensation. The apparatus still includes an encoder to encode the decoded picture or the decoded signal by motion-compensated prediction at the second accuracy of motion compensation using the second motion vector data to obtain a third bit stream, and a multiplexer to multiplex the third bit stream and the bit stream of the second motion vector data to obtain the second bit stream.
Furthermore, the present invention provides a method of converting a first bit stream of moving picture that has been encoded by motion-compensated prediction to a second bit stream of moving picture. A bit stream of first motion vector data is separated from the first bit stream. The first motion vector corresponds to first accuracy of motion compensation. A portion or all of the first bit stream is decoded by motion-compensated prediction at the first accuracy of motion compensation using the first motion vector data to obtain a decoded picture or a decoded signal at an intermediate processing stage. A bit stream of second motion vector data is formed by motion-compensated prediction using the first motion vector data. The second motion vector data corresponds to second accuracy of motion compensation that is
Le Vu
Victor Company of Japan Ltd.
LandOfFree
Conversion of video data bit stream does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Conversion of video data bit stream, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Conversion of video data bit stream will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2867280