Abstract: Systems and techniques are described herein for processing video data using a neural network system. For instance, a process can include generating, by a first convolutional layer of an encoder sub-network of the neural network system, output values associated with a luminance channel of a frame. The process can include generating, by a second convolutional layer of the encoder sub-network, output values associated with at least one chrominance channel of the frame. The process can include generating a combined representation of the frame by combining the output values associated with the luminance channel of the frame and the output values associated with the at least one chrominance channel of the frame. The process can include generating encoded video data based on the combined representation of the frame.

Abstract: This disclosure describes examples of extending the number of available discrete cosine transform (DCT) and discrete sine transform (DST) for encoding and decoding. A video coder may determine one or more transforms or inverse transforms to apply from a set of transforms or inverse transforms that includes DCT-2 or inverse DCT-2, DST-7 or inverse DST-7, DST-8 or inverse DST-8, DCT-3 or inverse DCT-3, DST-2 or inverse DST-2, DST-3 or inverse DST-3, DCT-4 or inverse DCT-4, DST-4 or inverse DST-4, DST-5 or inverse DST-5, DST-6 or inverse DST-6, and identity transform an inverse identity transform (IDT).

Abstract: An example device for coding video data includes a memory configured to store video data; and one or more processors implemented in circuitry and configured to: code a first codeword representing a selected transform scheme of a set of transform candidates of a multiple transform selection (MTS) scheme for a current block of video data, the selected transform scheme being a secondary transform of a set of available secondary transforms to be applied in addition to a primary transform; code a second codeword representing the secondary transform from the set of available secondary transforms; and apply the primary transform and the secondary transform during coding of residual data for the current block. The second codeword may be a value for a low-frequency non-separable transform (LFNST) syntax element.

Abstract: An example device for coding video data includes a memory configured to store video data; and one or more processors implemented in circuitry and configured to: code a first codeword representing a selected transform scheme of a set of transform candidates of a multiple transform selection (MTS) scheme for a current block of video data, the selected transform scheme being a secondary transform of a set of available secondary transforms to be applied in addition to a primary transform; code a second codeword representing the secondary transform from the set of available secondary transforms; and apply the primary transform and the secondary transform during coding of residual data for the current block. The second codeword may be a value for a low-frequency non-separable transform (LFNST) syntax element.

Abstract: Techniques are described for improving transform coding. For example, an encoded block of video data can be obtained, and a width and/or a height of the block can be determined. The width can be compared to a first threshold and/or the height can be compared to a second threshold. A horizontal transform and a vertical transform can be determined for the block based on comparing the width of the block to the first threshold and/or the height of the block to the second threshold. The horizontal transform and the vertical transform are determined without decoding a syntax element that indicates the horizontal transform and the vertical transform (e.g., the syntax element is not in an encoded video bitstream processed by a decoding device). In some cases, residual data is determined using the horizontal and vertical transforms, and a video block is determined using the residual data and a predictive block.

Abstract: An example method of decoding video data includes receiving encoded data for a current block. The method further includes determining whether single tree partitioning is enabled for the current block and decoding the encoded data to generate luma transform coefficients for the current block and chroma transform coefficients for the current block. The method further includes, in response to determining that the single tree partitioning is enabled for the current block, applying an inverse low-frequency non-separable transform (LFNST) to only the luma transform coefficients for the current block to produce a residual block for the current block. The method further includes reconstructing the current block of the video data using the residual block and a prediction block for the current block.

Abstract: An example device for decoding video data includes a memory configured to store the video data and one or more processors coupled to the memory. The one or more processors are configured to reorganize 2-D dequantized coefficients according to a first ordering. The one or more processors are configured to apply an inverse low-frequency non-separable transform (LFNST) to the reorganized 2-D dequantized coefficients to create inverse transformed coefficients. The one or more processors are configured to reorganize the inverse transformed coefficients according to a second ordering, the second ordering being based on an array including values, wherein each value in the array corresponds to a position in a 2-D block and the values in the array denote indices of the 2-D block in a defined order. The one or more processors are configured to decode the video data based on the second ordered inverse transformed coefficients.

Abstract: An example device for coding video data includes a memory configured to store the video data and one or more processors implemented in circuitry and communicatively coupled to the memory. The one or more processors are configured to determine whether scaling matrices may be applied to a low-frequency non-separable transform (LFNST) coded block. The one or more processors are also configured to, based on a determination that scaling matrices may not be applied to the LFNST coded block, not applying the scaling matrices to the LFNST coded block. The one or more processors are also configured to code the video data without applying the scaling matrices to the LFNST coded block.

Abstract: An example device for coding video data includes memory configured to store the video data and one or more processors implemented in circuitry and communicatively coupled to the memory. The one or more processors are configured to parse or signal all luma coefficients of a block of the video data from or to an encoded video bitstream. The one or more processors are configured to parse or signal at least one syntax element for the block after all the luma coefficients of the block are parsed or signaled from or to the encoded video bitstream, wherein the at least one syntax element comprises at least one of a low-frequency non-separable transform index for the luma coefficients or a multiple transform selection index. The one or more processors are also configured to code the block in accordance with the at least one syntax element.

Abstract: An example video codec includes memory configured to store the video data and one or more processors implemented in circuitry and communicatively coupled to the memory. The one or more processors are configured to determine that a current mode of coding a current block of the video data is a single tree partitioning mode. Based on the current mode being the single tree partitioning mode, the one or more processors are configured to refrain from determining whether there is a non-DC coefficient for a chroma component of a transform unit (TU) for the current block and refrain from coding a low-frequency non-separable transformation (LFNST) index in response to the refraining of the determination of whether there is the non-DC coefficient. The one or more processors are configured to code the current block in the single partitioning mode with LFNST disabled.

Abstract: A video decoder configured to determine a residual block by, for a first coefficient of a coefficient group, receiving a first instance of a flag, wherein the first instance of the flag is set to a first value; in response to the first instance of the flag being set to the first value, copying a coefficient value of a neighboring coefficient for the first coefficient; for a second coefficient of the coefficient group, receiving a second instance of the flag, wherein the second instance of the flag is set to a second value; in response to the second instance of the flag being set to the second value, receiving an index; determining, from a list of coefficient values, a value corresponding to the index; and set a value for the second coefficient to the value corresponding to the index.

Abstract: A video decoder may be configured to receive a block of video data, and determine a value of a multiple transform set index for the block of video data based on a presence of non-zero transform coefficients in the block of video data other than a DC coefficient. The video decoder may then apply a transform to the block of video data based on the determined value of the multiple transform set index.

Abstract: A video encoder and video decoder may code video data using intra prediction and a block-based delta pulse code modulation (BDPCM) mode. The BDPCM mode may include a vertical mode and a horizontal mode. The video encoder and video decoder may be configured to align the direction of an intra prediction mode to the direction of a BDPCM mode for both luma and chroma blocks.

Abstract: An example device for decoding video data includes a memory configured to store video data; and one or more processors implemented in circuitry and configured to: determine that a transform block of video data has a size of 8 8 coefficients and that the transform block is transformed using a low-frequency non-separable transform (LFNST); decode at least nine non-zero transform coefficients of the transform block; inverse transform the transform block using an inverse LFNST to reproduce a residual block corresponding to the transform block; and reconstruct a block of the video data using the residual block.

Abstract: An example device includes memory configured to store video data and one or more processors implemented in circuitry and coupled to the memory. The one or more processors determine whether a chroma block of the video data is encoded using dual tree partitioning. The one or more processors determine whether transform skip mode for the chroma block is enabled. The one or more processors, based on the chroma block being encoded using dual tree partitioning and transform skip mode being enabled for the chroma block, infer a value of a low-frequency non-separable transform (LFNST) index for the chroma block.

Abstract: A video coder may determine, for a transform block of video data, that at least one coefficient group, of the transform block, that comprises a non-zero transform coefficient is outside of a lowest frequency region of the transform block, wherein the at least one coefficient group is one of a plurality of coefficient groups that each comprise transform coefficients. The video coder may determine not to code a syntax element indicative of a multiple transform selection (MTS) for the transform block based at least in part on the determination of that the at least one coefficient group is outside of the lowest frequency region of the transform block. The video coder may code the video data based at least in part on the determination not to code the syntax element indicative of the multiple transform selection for the transform block.

Abstract: A video decoder is configured to determine a position of a last significant coefficient in a transform block of video data. The video decoder may then determine a value of a low-frequency non-separable transform (LFNST) index for the transform block based on the position of the last significant coefficient relative to a zero-out region of the transform block, wherein the zero-out region of the transform block includes both a first region within an LFNST region of the transform block and a second region of the transform block outside the LFNST region. The video decoder may then inverse transform the transform block in accordance with the value of the LFNST index.

Abstract: A video decoder can be configured to receive, in a syntax structure that applies to a current block, graph-related information; determine a transform matrix based on the received graph-related syntax information; perform an inverse transform based on the determined transform matrix of one or more coefficient values to generate a residual block; and reconstruct the current block of the video data based on the residual block.

Abstract: An example device for decoding video data includes a memory configured to store video data; and a processor implemented in circuitry and configured to decode a truncated unary codeword representing a multiple transform (MT) scheme for a current block of the video data to determine the MT scheme; apply the MT scheme to transform coefficients of the current block to produce residual data for the current block of video data; and decode the current block using the residual data. The MT scheme may include a plurality of transforms, such as a horizontal transform and a vertical transform, a primary transform and a second transform, or any combination of separable and/or non-separable transforms. Thus, a single truncated unary codeword may represent the entire MT scheme, that is, each of a plurality of transforms of the MT scheme.

Abstract: An example method of decoding video data that includes receiving one or more syntax elements of the video data indicative of whether a first type of coding scheme or a second type of coding scheme is applied to residual values of a block of video data coded with transform skip, wherein the residual values are indicative of a difference between the block and a prediction block, and wherein, in transform skip, the residual values are not transformed from a sample domain to a frequency domain. The method includes determining a type of coding scheme to apply to the residual values based on the one or more syntax elements, determining the residual values based on the determined type of coding scheme, and reconstructing the block based on the determined residual values and the prediction block.