Methods and apparatus for implicit adaptive motion vector predictor selection for video encoding and decoding

1. An apparatus, comprising:

a video encoder for encoding an image block using implicit adaptive motion vector predictor selection when at least one of spatio-temporal neighboring motion vector predictor candidates has a value that is inversely proportional to the motion compensation error caused by using the motion vector to compensate the corresponding block, and that is greater than or equal to a threshold, wherein a motion vector for encoding the block is predicted by defining a set of motion vector predictor candidates for the block based on a plurality of spatio-temporal neighboring motion vectors, and selection of a particular motion vector predictor candidate as the motion vector for encoding the block from the set is based on an estimated motion compensation error, motion vector magnitude, reference index, or other parameter available and related to motion information of the spatio-temporal neighboring motion vectors, and using explicit signaling of a motion vector predictor when none of said motion vector predictor candidates has said value that is greater than or equal to a threshold, wherein the respective values for the spatio-temporal neighboring motion vectors are used for a weighted median filter to derive the particular motion vector predictor candidate selected as the motion vector for encoding the block when implicit adaptive motion vector prediction selection is used, and wherein weights for the weighted median filter are a function of a fidelity value.

2. The apparatus of claim 1 , wherein the explicit signaling of motion vector predictors is used at a slice level, and the decision of whether to use the implicit adaptive motion vector predictor selection or the explicit signaling of motion vector predictors for the block is made at a block level.

3. In a video encoder, a method, comprising:

encoding an image block using implicit adaptive motion vector predictor selection when at least one of spatio-temporal neighboring motion vector predictor candidates has a value that is inversely proportional to the motion compensation error caused by using the motion vector to compensate the corresponding block, and that is greater than or equal to a threshold, wherein a motion vector for encoding the block is predicted by defining a set of motion vector predictor candidates for the block based on a plurality of spatio-temporal neighboring motion vectors, and selection of a particular motion vector predictor candidate as the motion vector for encoding the block from the set is based on an estimated motion compensation error, motion vector magnitude, reference index, or other parameter available and related to motion information of the spatio-temporal neighboring motion vectors, and using explicit signaling of a motion vector predictor when none of said motion vector predictor candidates has said value that is greater than or equal to a threshold, wherein the respective values for the spatio-temporal neighboring motion vectors are used for a weighted median filter to derive the particular motion vector predictor candidate selected as the motion vector for encoding the block when implicit adaptive motion vector prediction selection is used, and wherein weights for the weighted median filter are a function of a fidelity value.

4. The method of claim 3 , wherein the explicit signaling of motion vector predictors is used at a slice level, and the decision of whether to use the implicit adaptive motion vector predictor selection or the explicit signaling of motion vector predictors for the block is made at a block level.

5. An apparatus, comprising:

a video decoder for decoding an image block using implicit adaptive motion vector predictor selection when at least one of spatio-temporal neighboring motion vector predictor candidates has a value that is inversely proportional to the motion compensation error caused by using the motion vector to compensate the corresponding block, and that is greater than or equal to a threshold, wherein a motion vector for decoding the block is predicted by defining a set of motion vector predictor candidates for the block based on a plurality of spatio-temporal neighboring motion vectors, and selection of a particular motion vector predictor candidate as the motion vector for decoding the block from the set is based on an estimated motion compensation error, motion vector magnitude, reference index, or other parameter available and related to motion information of the spatio-temporal neighboring motion vectors, and using explicit signaling of a motion vector predictor when none of said motion vector predictor candidates has said value that is greater than or equal to a threshold, wherein the respective values for the spatio-temporal neighboring motion vectors are used for a weighted median filter to derive the particular motion vector predictor candidate selected as the motion vector for encoding the block when implicit adaptive motion vector prediction selection is used, and wherein weights for the weighted median filter are a function of a fidelity value.

6. The apparatus of claim 5 , wherein the explicit signaling of motion vector predictors is used at a slice level, and the decision of whether to use the implicit adaptive motion vector predictor selection or the explicit signaling of motion vector predictors for the block is made at a block level.

7. In a video decoder, a method, comprising:

decoding an image block using implicit adaptive motion vector predictor selection when at least one of spatio-temporal neighboring motion vector predictor candidates has a value that is inversely proportional to the motion compensation error caused by using the motion vector to compensate the corresponding block, and that is greater than or equal to a threshold, wherein a motion vector for decoding the block is predicted by defining a set of motion vector predictor candidates for the block based on a plurality of spatio-temporal neighboring motion vectors, and selection of a particular motion vector predictor candidate as the motion vector for decoding the block from the set is based on an estimated motion compensation error, motion vector magnitude, reference index, or other parameter available and related to motion information of the spatio-temporal neighboring motion vectors, and using explicit signaling of a motion vector predictor when none of said motion vector predictor candidates has said value that is greater than or equal to a threshold, wherein the respective values for the spatio-temporal neighboring motion vectors are used for a weighted median filter to derive the particular motion vector predictor candidate selected as the motion vector for encoding the block when implicit adaptive motion vector prediction selection is used, and wherein weights for the weighted median filter are a function of a fidelity value.

8. The method of claim 7 , wherein the explicit signaling of motion vector predictors is used at a slice level, and the decision of whether to use the implicit adaptive motion vector predictor selection or the explicit signaling of motion vector predictors for the block is made at a block level.

9. A computer readable non-transitory storage media having instructions that when executed by one or more processors cause the one or more processors to perform operations for

encoding video signal data thereupon, and

compressing data for an image block encoded using implicit adaptive motion vector predictor selection when at least one of spatio-temporal neighboring motion vector predictor candidates has a value that is inversely proportional to the motion compensation error caused by using the motion vector to compensate the corresponding block, and that is greater than or equal to a threshold,

wherein a motion vector for encoding the block is predicted by defining a set of motion vector predictor candidates for the block based on a plurality of spatio-temporal neighboring motion vectors, and selection of a particular motion vector predictor candidate as the motion vector for encoding the block from the set is based on an estimated motion compensation error, motion vector magnitude, reference index, or other parameter available and related to motion information of the spatiotemporal neighboring motion vectors, and using explicit signaling of a motion vector predictor when none of said motion vector predictor candidates has said value that is greater than or equal to a threshold, wherein the respective values for the spatio-temporal neighboring motion vectors are used for a weighted median filter to derive the particular motion vector predictor candidate selected as the motion vector for encoding the block when implicit adaptive motion vector prediction selection is used, and wherein weights for the weighted median filter are a function of a fidelity value.