Motion vector coding for video coding
Coding a motion vector difference (MVD) during an inter-prediction process. Example techniques may include determining a particular coding and/or signaling method for an MVD from among two or more MVD coding and/or signaling techniques. A video coder (e.g., a video encoder and/or a video decoder) may determine a particular MVD coding and/or signaling technique based on characteristics of video data or coding methods, including MV precision, Picture Order Count (POC) difference, or any other already coded/decoded information of a block of video data.
1. A method of decoding video data, the method comprising:
receiving an encoded block of video data;
receiving one or more syntax elements indicating a motion vector difference (MVD) associated with the encoded block of video data;
determining a set of context models from two or more sets of context models used to entropy code the one or more syntax elements indicating the MVD, the set of context models being determined based on a motion model for the encoded block of video data being an affine motion model or a translational motion model;
entropy decoding the one or more syntax elements indicating the MVD using the determined set of context models to determine a decoded MVD; and
decoding the encoded block of video data using the decoded MVD.
2. The method of claim 1 , wherein determining the set of context models from the two or more sets of context models used to entropy code the one or more syntax elements indicating the MVD comprises:
determining the set of context models from the two or more sets of context models based on the motion model and one or more of a block size of the encoded block of video data, a motion vector precision used for the MVD, an inter prediction direction, a reference picture, a picture order count difference between the encoded block of video data and the reference picture, a motion vector of a neighboring block, an MVD of a neighboring block, intra/inter mode decision of neighboring blocks, or a motion predictor characteristic.
3. An apparatus configured to decode video data, the apparatus comprising:
a memory configured to store an encoded block of video data; and
one or more processors implemented in circuitry and in communication with the memory, the one or more processors configured to:
receive an encoded block of video data;
receive one or more syntax elements indicating a motion vector difference (MVD) associated with the encoded block of video data;
determine a set of context models from two or more sets of context models used to entropy code the one or more syntax elements indicating the MVD, the set of context models being determined based on a motion model for the encoded block of video data being an affine motion model or a translational motion model;
entropy decode the one or more syntax elements indicating the MVD using the determined set of context models to determine a decoded MVD; and
decode the encoded block of video data using the decoded MVD.
4. The apparatus of claim 3 , wherein to determine the set of context models from the two or more sets of context models used to entropy code the one or more syntax elements indicating the MVD, the one or more processors are configured to:
determine the set of context models from the two or more sets of context models based on the motion model and one or more of a block size of the encoded block of video data, a motion vector precision used for the MVD, an inter prediction direction, a reference picture, a picture order count difference between the encoded block of video data and the reference picture, a motion vector of a neighboring block, an MVD of a neighboring block, intra/inter mode decision of neighboring blocks, or a motion predictor characteristic.
5. An apparatus configured to decode video data, the apparatus comprising:
means for receiving an encoded block of video data;
means for receiving one or more syntax elements indicating a motion vector difference (MVD) associated with the encoded block of video data;
means for determining a set of context models from two or more sets of context models used to entropy code the one or more syntax elements indicating the MVD, the set of context models being determined based on a motion model for the encoded block of video data being an affine motion model or a translational motion model;
means for entropy decoding the one or more syntax elements indicating the MVD using the determined set of context models to determine a decoded MVD; and
means for decoding the encoded block of video data using the decoded MVD.
6. The apparatus of claim 5 , wherein the means for determining the set of context models from the two or more sets of context models used to entropy code the one or more syntax elements indicating the MVD comprises:
means for determining the set of context models from the two or more sets of context models based on the motion model and one or more of a block size of the encoded block of video data, a motion vector precision used for the MVD, an inter prediction direction, a reference picture, a picture order count difference between the encoded block of video data and the reference picture, a motion vector of a neighboring block, an MVD of a neighboring block, intra/inter mode decision of neighboring blocks, or a motion predictor characteristic.
7. A non-transitory computer-readable storage medium storing instructions that, when executed, cause one or more processors of a device configured to decode video data to:
receive an encoded block of video data;
receive one or more syntax elements indicating a motion vector difference (MVD) associated with the encoded block of video data;
determine a set of context models from two or more sets of context models used to entropy code the one or more syntax elements indicating the MVD, the set of context models being determined based on a motion model for the encoded block of video data being an affine motion model or a translational motion model;
entropy decode the one or more syntax elements indicating the MVD using the determined set of context models to determine a decoded MVD; and
decode the encoded block of video data using the decoded MVD.
8. The non-transitory computer-readable storage medium of claim 7 , wherein to determine the set of context models from the two or more sets of context models used to entropy code the one or more syntax elements indicating the MVD, the instructions further cause the one or more processors to:
determine the set of context models from the two or more sets of context models based on the motion model and one or more of a block size of the encoded block of video data, a motion vector precision used for the MVD, an inter prediction direction, a reference picture, a picture order count difference between the encoded block of video data and the reference picture, a motion vector of a neighboring block, an MVD of a neighboring block, intra/inter mode decision of neighboring blocks, or a motion predictor characteristic.