Method and apparatus for video coding with primary and secondary transforms

An apparatus, comprising:

• processing circuitry configured to:
• determine an intra prediction mode for generating a prediction image of a current block;
• determine a pair of transforms including a horizontal transform and a vertical transform that are transforms of a default type for performing a primary transform process for the current block when a first flag extracted from a coded video bitstream has a first value;
• determine that a secondary transform process is not to be performed for the current block when the first flag extracted from the coded video bitstream has the first value;
• determine the pair of transforms for performing the primary transform process based on a transform index extracted from the coded video bitstream when the first flag extracted from the coded video bitstream has a second value;
• determine that the secondary transform process is to be performed for the current block when the first flag extracted from the coded video bitstream has the second value;
• when the secondary transform process is determined to be performed,
• determine a secondary transform for performing the secondary transform process according to the determined intra prediction mode and the determined pair of transforms, and
• reconstruct a residue image of the current block based on one or more coefficients of an input block extracted from the coded video bitstream, the determined pair of transforms, and the determined secondary transform;
• when the secondary transform process is determined not to be performed,
• reconstruct the residue image of the current block based on the one or more coefficients of the input block and the determined pair of transforms;
• and
• reconstruct an image of the current block based on the prediction image and the residue image of the current block.

A non-transitory computer-readable medium storing instructions which when executed by a computer for video decoding causes the computer to perform:

• determining an intra prediction mode for generating a prediction image of a current block;
• determining a pair of transforms including a horizontal transform and a vertical transform that are transforms of a default type for performing a primary transform process for the current block when a first flag extracted from a coded video bitstream has a first value;
• determining that a secondary transform process is not to be performed for the current block when the first flag extracted from the coded video bitstream has the first value;
• determining the pair of transforms for performing the primary transform process based on a transform index extracted from the coded video bitstream when the first flag extracted from the coded video bitstream has a second value;
• determining that the secondary transform process is to be performed for the current block when the first flag extracted from the coded video bitstream has the second value;
• when the secondary transform process is determined to be performed,
• determining a secondary transform for performing the secondary transform process according to the determined intra prediction mode and the determined pair of transforms, and
• reconstructing a residue image of the current block based on one or more coefficients of an input block extracted from the coded video bitstream, the determined pair of transforms, and the determined secondary transform;
• when the secondary transform process is determined not to be performed,
• reconstructing the residue image of the current block based on the one or more coefficients of the input block and the determined pair of transforms;
• and
• reconstructing an image of the current block based on the prediction image and the residue image of the current block.