Mode-dependent joint component transform
A method for coding video data, executable by a processor, includes receiving video data; entropy-parsing the received video data into one or more components; de-quantizing the one or more entropy-parsed components; performing a joint component secondary transformation (JCST) on the one or more components in accordance with a JCST kernel selected based on one of (i) a prediction mode corresponding to the video data, (ii) a primary transform type of a current block, and (iii) a secondary transform kernel selected for the current block; and decoding the video data based on one or more residual components corresponding to the joint component secondary transformed components.
1. A method for coding video data, executable by a processor, the method comprising:
receiving video data;
entropy-parsing the received video data into one or more components;
de-quantizing the one or more entropy-parsed components;
performing a joint component secondary transformation (JCST) on the one or more components in accordance with a JCST kernel selected based on one of (i) a prediction mode corresponding to the video data, (ii) a primary transform type of a current block, and (iii) a secondary transform kernel selected for the current block; and
decoding the video data based on one or more residual components corresponding to the joint component secondary transformed components.
2. The method of claim 1 , wherein the JCST kernel is selected based on the prediction mode corresponding to the video data.
3. The method of claim 2 , wherein the JCST kernel is applied based on an intra prediction mode associated with the video data.
4. The method of claim 3 , wherein the intra prediction mode comprises one or more from among DC, SMOOTH_H, SMOOTH_V, directional modes, Paeth predictor, and filtering modes.
5. The method of claim 3 , wherein the JCST kernel is applied based on a nominal intra prediction modes.
6. The method of claim 3 , wherein JCST is disabled based on the intra prediction mode.
7. The method of claim 6 , wherein the intra prediction mode comprises one or more from among chroma-from-luma mode and Paeth Predictor.
8. The method of claim 2 , wherein different JCST kernels are applied for intra and inter prediction modes.
9. The method of claim 2 , wherein the JCST kernel is applied based on an Intra Block Copy mode.
10. The method of claim 2 , wherein the JCST kernel is applied based on a Palette mode.
11. The method of claim 2 , wherein the JCST kernel is applied based on different inter prediction modes.
12. The method of claim 2 , wherein the JCST is applied based on motion information.
13. The method of claim 12 , wherein the JCST kernels is dependent on whether a motion vector predictor comes from a top, a left, or a temporal motion vector predictor candidate.
14. The method of claim 12 , wherein the JCST kernel is dependent on a magnitude of the motion vector.
15. The method of claim 12 , wherein the JCST kernel is dependent on a direction of the motion vector.
16. The method of claim 1 , wherein the JCST kernel is selected based on the primary transform type selected for the current block associated with the video data.
17. The method of claim 16 , wherein the JCST kernel is applied based on a primary transform being applied on color components associated with the video data applying JCST.
18. The method of claim 1 , wherein the JCST kernel is selected based on the secondary transform kernel selected for the current block associated with the video data.
19. A decoder for decoding video data, the computer system comprising:
one or more computer-readable non-transitory storage media configured to store computer program code; and
one or more computer processors configured to access said computer program code and operate as instructed by said computer program code, said computer program code including:
receiving code configured to cause the one or more computer processors to receive video data;
entropy-parsing code configured to cause the one or more computer processors to entropy-parse the received video data into one or more components;
de-quantizing code configured to cause the one or more computer processors to de-quantize the one or more entropy-parsed components;
performing code configured to cause the one or more computer processors to perform a joint component secondary transformation (JCST) on the one or more components in accordance with a JCST kernel selected based on one of (i) a prediction mode corresponding to the video data, (ii) a primary transform type of a current block, and (iii) a secondary transform kernel selected for the current block; and
decoding code configured to cause the one or more computer processors to decode the video data based on one or more residual components corresponding to the joint component secondary transformed components.
20. A method for decoding a video bitstream, executable by a processor, the method comprising:
decoding the video bitstream based on one or more residual components corresponding to joint component secondary transformed components,
wherein the video bitstream is entropy-parsed into one or more components;
wherein the one or more entropy-parsed components are de-quantized;
wherein a joint component secondary transformation (JCST) is performed on the one or more de-quantized components in accordance with a JCST kernel selected based on one of (i) a prediction mode corresponding to the video data, (ii) a primary transform type of a current block, and (iii) a secondary transform kernel selected for the current block.