FIELD: video encoding.
SUBSTANCE: invention concerns the ability to implicitly select a transformation when encoding video. To do this, the method includes inferring a transform type from a plurality of transform types, which includes one or more discrete cosine transforms (DCT) and one or more discrete sine transforms (DST), wherein the inference comprises: determining the size of the current transform block; determining whether or not the current video block is segmented using internal subblock segmentation (ISP); and in response to determining that the size of the current transform block is less than a threshold and that the current video block is segmented using the ISP, selecting a particular DST from the one or more DSTs as the selected transform type; transforming, using the selected transform type, the current transform block to obtain a reconstructed residual data block for the video block; and recovering the video block based on the recovered residual data for the video block.
EFFECT: ability to implicitly select a transformation when encoding video.
7 cl, 13 dwg, 1 tbl
Title | Year | Author | Number |
---|---|---|---|
TREE-BASED TRANSFORMATION UNIT (TU) SEGMENTATION FOR VIDEO ENCODING | 2019 |
|
RU2809181C2 |
BLOCK ADAPTIVE CODING WITH CONVERSION OF COLOUR SPACE | 2015 |
|
RU2698760C2 |
AFFINE LINEAR WEIGHTED INTERNAL PREDICTION IN VIDEO ENCODING | 2020 |
|
RU2816748C2 |
BLOCK-DIFFERENTIAL PULSE-CODE MODULATION IN FIELD OF COEFFICIENTS DURING VIDEO ENCODING | 2020 |
|
RU2825319C2 |
REDUCTION OF THE NUMBER OF REGULARLY-CODED RESOLUTION ELEMENTS FOR DECODING COEFFICIENTS USING THRESHOLD VALUE AND RICE PARAMETER | 2019 |
|
RU2793272C2 |
CODING FLAGS OF CODED BLOCKS | 2013 |
|
RU2627119C2 |
CODING COEFFICIENTS FOR SKIP CONVERSION MODE | 2020 |
|
RU2813008C2 |
INTRA-INTER PREDICTION MODE FOR VIDEO DATA | 2020 |
|
RU2803896C2 |
DETERMINATION OF CONTEXTS FOR DATA CODING OF TRANSFORMATION COEFFICIENTS IN VIDEO CODING | 2013 |
|
RU2595897C2 |
PREDICTION IN INTERNAL DC MODE WHEN ENCODING VIDEO | 2020 |
|
RU2817294C2 |
Authors
Dates
2023-11-21—Published
2020-03-12—Filed