Decoding method and decoder with rounding means
View Patent ↗The present invention relates to a video decoder for decoding a bit stream corresponding to pictures of a video signal. The invention is such that, motion vectors having values coded at a nominal resolution in the bit stream, said video decoder includes a rounding unit for rounding decoded nominal motion vectors to a resolution different from the nominal resolution in a way that minimizes the error accumulation along prediction path by assigning to rounded successive motion vectors along the prediction path, successive values such that their average value is equal to the value of the nominal motion vector.
1. A video decoder for decoding a bit stream corresponding to pictures of a video signal, characterized by motion vectors having values coded at a nominal resolution in the bit stream, said video decoder comprising:
a signal-input line configured to receive the bit stream;
means for rounding decoded nominal motion vectors to a resolution different from the nominal resolution to lessen the error accumulation along a prediction path by assigning to rounded successive motion vectors along the prediction path, successive values such that the average value of the rounded motion vectors is equal to the value of the nominal motion vector.
2. A video decoder as claimed in claim 1 , in which said means for rounding nominal motion vectors alternates at least two motion vector rounding policies from one picture to the other picture of the same type.
3. A video decoder as claimed in claim 1 , in which said means for rounding nominal motion vectors alternates at least two motion vector rounding policies from one reference picture to the other reference picture.
4. A mobile device including a video decoder as claimed in claim 1 .
5. The video decoder of claim 1 , wherein the bit stream is a MPEG-4 bit stream.
6. The video decoder of claim 1 , wherein the bit stream is a DivX bit stream.
7. A method for decoding a bit stream corresponding to pictures of a video signal, characterized by motion vectors having values coded at a nominal resolution in the bit stream, the decoding method comprising:
rounding decoded nominal motion vectors to a resolution different from the nominal resolution in a way that minimizes the error accumulation along prediction path by and including assigning to rounded successive motion vectors along the prediction path, successive values such that the average value of the rounded motion vectors is equal to the value of the nominal motion vector.
8. A non-transitory computer program product comprising program instructions for implementing, when said program is executed by a processor, a decoding method as claimed in claim 7 .
9. The method of claim 7 , wherein the bit stream is a MPEG-4 bit stream.
10. The method of claim 7 , wherein the bit stream is a DivX bit stream.
11. A video decoder for decoding a bit stream corresponding to pictures of a video signal, characterized by motion vectors having values coded at a nominal resolution in the bit stream, the video decoder comprising:
a signal-input line configured to receive the bit stream;
a rounding circuit configured and arranged to decoded nominal motion vectors to a resolution different from the nominal resolution by using at least two motion vector rounding policies, the at least two rounding policies providing sufficient video quality by rounding the motion vectors to values, including fractional values, to lessen the error accumulation along a prediction path by assigning to rounded successive motion vectors along the prediction path successive values such that the average value of the rounded motion vectors is equal to the value of the nominal motion vector.
12. A method for decoding a bit stream corresponding to pictures of a video signal, characterized by motion vectors having values coded at a nominal resolution in the bit stream, the decoding method comprising:
rounding decoded nominal motion vectors to a resolution different from the nominal resolution by
using at least two motion vector rounding policies, the at least to rounding policies providing sufficient video quality by rounding the motion vectors to fractional integer values in a way that lessens the error accumulation along prediction path by assigning to rounded successive motion vectors along the prediction path successive values such that the average value of the rounded motion vectors is equal to the value of the nominal motion vector.