Back to results
US 12,368,850 B2Granted
Method of signaling in a video codec
AssigneeBritish Broadcasting Corporation
InventorsSaverio Blasi, Andre Seixas Dias, Gosala Kulupana
Life of this patent
Estimating the term…
Abstract
Encoding of video data in a video codec involves a transform of residuals. This can be composed of a primary transform and a secondary transform. The selection of the secondary transform is effected by considering the characteristics of a block to be encoded. The selection of secondary transform can be signaled to the decoder, or inferred therein.
Claims15
A decoder for decoding an encoded bitstream representative of a block of a frame of video, the decoder comprising:
- an inverse transform module comprising executable instructions that cause one or more processors to:
- determine a total number of non-zero coefficients contained in the block to be decoded,
- determine a set of candidate inverse secondary transform matrices on the basis of the determined total number of non-zero coefficients contained in the block, wherein the set of candidate inverse secondary transform matrices comprises a plurality of inverse secondary transform matrices, and wherein determining the set of candidate inverse secondary transform matrices comprises determining a candidate number on the basis of the determined total number of non-zero coefficients contained in the block, the candidate number determining how many candidate inverse secondary transform matrices are to be determined in the set of candidate inverse secondary transform matrices,
- select an inverse secondary transform matrix from the set of candidate inverse secondary transform matrices based on a signal received on the encoded bitstream, and
- apply an inverse matrix transformation to transformed residual information, to extract untransformed residual information, the inverse matrix transformation being governed by the inverse secondary transform matrix and an inverse primary transform matrix;
- and
- an intra-prediction module comprising executable instructions that cause the one or more processors to compute a prediction of the block in accordance with an intra-prediction mode and reconstructing the block by combining the inverse transformed residual data with the prediction.
4 more elements
A method of decoding encoded transformed residual information for a block of a frame of video, the method comprising:
- determining a total number of non-zero coefficients contained in the block to be decoded;
- determining a set of candidate inverse secondary transform matrices on the basis of the total number of non-zero coefficients contained in the block to be decoded, wherein the set of candidate inverse secondary transform matrices comprises a plurality of inverse secondary transform matrices, and determining the set of candidate inverse secondary transform matrices comprises determining a candidate number on the basis of the determined total number of non-zero coefficients contained in the block, the candidate number determining how many candidate inverse secondary transform matrices are to be determined in the set of candidate inverse secondary transform matrices;
- selecting an inverse secondary transform matrix from the set of candidate inverse secondary transform matrices based on a signal received on a bitstream;
- inverse transforming the residual information applying an inverse secondary transform and an inverse primary transform, to extract untransformed residual information, wherein the inverse secondary transform is computed on the basis of the inverse secondary transform matrix;
- computing a prediction of the current block in accordance with an intra-prediction mode;
- and
- reconstructing the block by combining the inverse transformed residual data with the prediction.
4 more elements
Dependencies are inferred from the claim text.