Methods for efficent application of lgt
AssigneeTencent America LLC
InventorsMadhu Peringassery KRISHNA, Xin ZHAO, Shan LIU
Life of this patent
Estimating the term…
Abstract
A method of encoding image data performed by at least one processor, may include: receiving information regarding a data block of an image; performing control so that a residual block composed of residual signals after intra or inter prediction is generated; identifying one or more hybrid transform kernel that are available for encoding the residual block based on associations between prediction modes and sizes of residual blocks; selecting a hybrid transform kernel from among the one or more hybrid transform kernel that are available for encoding the residual block; and performing transform coding of the residual block using the selected hybrid transform kernel.
Claims20
A method of encoding image data performed by at least one processor, the method comprising:
- receiving information regarding a data block of an image;
- performing control so that a residual block composed of residual signals after intra or inter prediction is generated;
- identifying one or more hybrid transform kernel that are available for encoding the residual block based on associations between prediction modes and sizes of residual blocks, wherein each of the one or more hybrid transform kernel includes at least two transform types, which includes at least two transform types that are the same, or at least two transform types that are different;
- selecting a hybrid transform kernel from among the one or more hybrid transform kernel that are available for encoding the residual block;
- and
- performing transform coding of the residual block using the selected hybrid transform kernel.
An apparatus comprising:
- at least one memory configured to store computer program code;
- and
- at least one processor configured to access the at least one memory and operate according to the computer program code, the computer program code comprising:
- receiving code configured to cause the at least one processor to receive information regarding a data block of an image;
- residual block generation code configured to cause the at least one processor to perform control so that a residual block composed of residual signals after intra or inter prediction is generated;
- identifying code configured to cause the at least one processor to identify one or more hybrid transform kernel that are available for encoding the residual block based on associations between prediction modes and sizes of residual blocks, wherein each of the one or more hybrid transform kernel includes at least two transform types, which includes at least two transform types that are the same, or at least two transform types that are different;
- selecting code configured to cause the at least one processor to selecting a hybrid transform kernel from among the one or more hybrid transform kernel that are available for encoding the residual block;
- and
- transform coding code configured to cause the at least one processor to perform transform coding of the residual block using the selected hybrid transform kernel.
A method of decoding the bitstream composed of video/image data performed by at least one processor, the method comprising:
- receiving a bitstream composed of compressed video/image data;
- generating dequantized blocks by a process including but not limited to parsing the bitstream, decoding the entropy coded bitstream and dequantizing each block;
- identifying hybrid transform kernels that are available based on associations between prediction modes and sizes of blocks;
- selecting a hybrid transform kernel from among the hybrid transform kernels that are available for decoding;
- and
- performing transform coding of the dequantized blocks using inverse of transforms in the hybrid transform kernel.
Dependencies are inferred from the claim text.