Method and apparatus for processing image signal

A method for encoding an image signal, comprising:

• determining an input length and an output length of a non-separable transform based on a size of a current block;
• determining a non-separable transform matrix based on an intra prediction mode of a current block;
• and
• applying the non-separable transform matrix to coefficients of the current block, wherein the input length and the output length of the non-separable transform are respectively determined as 8 and 16 based on the size of the current block being 4×4.

A non-transitory decoder-readable medium for storing a bitstream generated by a video encoder, the bitstream comprising:

• information used for determining an input length and an output length of a non-separable transform based on a size of a current block;
• and
• information used for determining a non-separable transform matrix based on an intra prediction mode of a current block,
• wherein the input length and the output length of the non-separable transform are respectively determined as 8 and 16, based on the size of the current block being 4×4.