Inter modes with weighting-averaged prediction, symmetric mvd or affine motion in video coding

A device for coding video data, the device comprising:

• a memory to store the video data;
• and
• one or more processors implemented in circuitry, the one or more processors configured to:
• form, in a symmetric motion vector difference mode, a List 0 (L0) base vector using a L0 Advanced Motion Vector Prediction (AMVP) candidate list and form a List 1 (L1) base vector using a L1 AMVP candidate list;
• determine a refined L0 motion vector and a refined L1 motion vector by performing a decoder-side motion vector refinement process that refines the L0 base vector and the L1 base vector, wherein the refined L0 motion vector indicates a location in an L0 reference picture in a reference picture list L0 and the refined L1 motion vector indicates a location in an L1 reference picture in a reference picture list L1;
• use the refined L0 motion vector and the refined L1 motion vector to determine a prediction block for a current coding unit (CU) of a current picture of the video data;
• inverse quantize a transform coefficient block, wherein the transform coefficient block includes a DC transform coefficient;
• apply an inverse transform to the inverse quantized transform coefficient block to generate residual data;
• prior to reconstructing the current CU based on the prediction block, add an offset value to prediction samples of the prediction block of the CU, wherein the offset value is smaller than a quantization step size of the DC transform coefficient;
• and
• reconstruct the current CU based on the prediction block by combining the prediction block and the residual data.

A computer-readable storage medium having stored thereon instructions that, when executed, cause one or more processors to:

• form, in a symmetric motion vector difference mode, a List 0 (L0) base vector using a L0 Advanced Motion Vector Prediction (AMVP) candidate list and form a List 1 (L1) base vector using a L1 AMVP candidate list;
• determine a refined L0 motion vector and a refined L1 motion vector by performing a decoder-side motion vector refinement process that refines the L0 base vector and the L1 base vector, wherein the refined L0 motion vector indicates a location in an L0 reference picture in a reference picture list L0 and the refined L1 motion vector indicates a location in an L1 reference picture in a reference picture list L1;
• use the refined L0 motion vector and the refined L1 motion vector to determine a prediction block for a current coding unit (CU) of a current picture of video data;
• inverse quantize a transform coefficient block, wherein the transform coefficient block includes a DC transform coefficient;
• apply an inverse transform to the inverse quantized transform coefficient block to generate residual data;
• prior to reconstructing the current CU based on the prediction block, add an offset value to prediction samples of the prediction block of the CU, wherein the offset value is smaller than a quantization step size of the DC transform coefficient;
• and
• reconstruct the current CU based on the prediction block by combining the prediction block and the residual data.