Affine motion compensation for current picture referencing

1. An apparatus comprising:

one or more processors for processing of data;

a memory for storage of data including video data; and

an encoder to encode video data to generate encoded video data and provide affine motion compensation for current picture references in the video data, wherein the encoded video data includes data relating to cube projections for the video data including cube face locations, discontinuities, and rotations in a cube projection such that the encoder to track the cube face locations, the discontinuities, and the rotations in the cube projection for selection of one or more search neighborhoods for motion estimation for affine control point motion vectors associated with the video data, wherein the encoder to determine a block of the video data eligible for the affine motion compensation, wherein determining eligibility for affine motion compensation includes determining one or more of whether one or more pixel positions of the block are located in an invalid area or whether one or more sub-blocks of the block are located in an invalid area.

2. The apparatus of claim 1 , wherein the encoding of the video data is to provide sub-pel motion vector resolution for a block of the video data.

3. The apparatus of claim 2 , wherein the encoder includes a locally adaptive motion vector resolution (LAMVR) component, wherein the encoder is to utilize the LAMVR component to avoid sub-pel filter computations that use pixels from discontinuous faces of a cube face.

4. The apparatus of claim 1 , further comprising a decoder to decode the encoded video data, the decoder including a component to decode affine motion compensation for current picture references in the encoded video data, wherein the decoder further includes data regarding cube projections, including cube face locations, discontinuities, and rotations in a cube projection in the encoded video data.

5. One or more non-transitory computer-readable storage mediums having stored thereon data representing sequences of instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:

receiving video data for encoding;

encoding the video data utilizing a video encoder, wherein the encoding of the video data includes applying affine motion compensation for current picture references in the video data, wherein encoding the video data includes application of knowledge data relating to cube projections in the video data, including cube face locations, discontinuities, and rotations in a cube projection such that the encoder to track the cube face locations, the discontinuities, and the rotations in the cube projection for selection of one or more search neighborhoods for motion estimation for affine control point motion vectors associated with the video data;

determining a block in the video data eligible for affine motion compensation, wherein determining eligibility for affine motion compensation includes determining one or more of whether one or more pixel positions of the block are located in an invalid area or whether one or more sub-blocks of the block are located in an invalid area; and

outputting encoded video data.

6. The one or more storage mediums of claim 5 , wherein determining eligibility for affine motion compensation includes determining whether one or more pixel positions of the block are located in an invalid area.

7. The one or more storage mediums of claim 5 , wherein determining eligibility for affine motion compensation includes determining whether one or more sub-blocks of the block are located in an invalid area.

8. The one or more storage mediums of claim 5 , wherein the encoding of the video data includes providing sub-pel motion vector resolution for a block in the video data.

9. The one or more storage mediums of claim 8 , wherein the operations further comprise:

utilizing a locally adaptive motion vector resolution (LAMVR) component to avoid sub-pel filter computations that use pixels from discontinuous faces of a cube face in the video data.

10. The one or more storage mediums of claim 5 , wherein the operations further comprise facilitating decoding the encoded video data, wherein the decoding of the video data includes decoding affine motion compensation for current picture references in the video data, wherein decoding the video data includes application of knowledge data regarding cube projections in the video data, including cube face locations, discontinuities, and rotations in a cube projection.

11. A method comprising:

receiving video data for encoding;

encoding the video data utilizing a video encoder, wherein the encoding of the video data includes applying affine motion compensation for current picture references in the video data, wherein encoding the video data includes application of knowledge data relating to cube projections in the video data, including cube face locations, discontinuities, and rotations in a cube projection such that the encoder to track the cube face locations, the discontinuities, and the rotations in the cube projection for selection of one or more search neighborhoods for motion estimation for affine control point motion vectors associated with the video data;

determining a block in the video data eligible for affine motion compensation, wherein determining eligibility for affine motion compensation includes determining one or more of whether one or more pixel positions of the block are located in an invalid area or whether one or more sub-blocks of the block are located in an invalid area; and

outputting encoded video data.

12. The method of claim 11 , wherein the encoding of the video data includes providing sub-pel motion vector resolution for a block in the video data.

13. The method of claim 11 , further comprising:

utilizing a locally adaptive motion vector resolution (LAMVR) component to avoid sub-pel filter computations that use pixels from discontinuous faces of a cube face in the video data.

14. The method of claim 11 , further comprising facilitating decoding of the encoded video data, wherein the decoding of the video data includes decoding affine motion compensation for current picture references in the video data, wherein decoding the video data includes application of knowledge data regarding cube projections in the video data, including cube face locations, discontinuities, and rotations in a cube projection.