LPDXVA_HuffmanTable_MJPEG - NtDoc
Native API online documentation, based on the System Informer
(formerly Process Hacker) phnt
headers
// dxva.h
typedef struct _DXVA_HuffmanTable_MJPEG {
UINT8 bits_ac[4][16];
UINT8 table_ac[4][256];
UINT8 bits_dc[4][16];
UINT8 table_dc[4][256];
} DXVA_HuffmanTable_MJPEG, *LPDXVA_HuffmanTable_MJPEG;
View the official Windows Driver Kit DDI referenceNo description available.
The DXVA_HuffmanTable_MJPEG structure provides the Huffman table of a compressed picture for MJPEG video decoding.
bits_ac[4]The bits_ac[j][i] member specifies the number of Huffman codes of length i for Huffman table slot j for AC class tables. This value is the same as the Li parameter of the DHT marker from the specification.
table_ac[4]The table_ac[j][i] member specifies the value associated with each Huffman code j for Huffman table slot I for AC class tables. This corresponds to the Vi,j parameter from the DHT marker of the specification.
bits_dc[4]The bits_dc[j][i] member specifies the number of Huffman codes of length i for Huffman table slot j for DC class tables. This value is the same as the Li parameter of the DHT marker from the specification.
table_dc[4]The table_dc[j][i] member specifies the value associated with each Huffman code j for Huffman table slot I for DC class tables. This corresponds to the Vi,j parameter from the DHT marker of the specification.
This structure is used for MJPEG video coding when the buffer type is D3D12_VIDEO_DECODE_ARGUMENT_TYPE_HUFFMAN_TABLE in D3D12 and D3D11_VIDEO_DECODER_BUFFER_HUFFMAN_TABLE in D3D11.
If DXVA_HuffmanTable_MJPEG is left unspecified by the host, the accelerator shall use the default quantization table as specified in Annex K of the JPEG specification.
D3D12_VIDEO_DECODE_ARGUMENT_TYPE
D3D11_VIDEO_DECODER_BUFFER_TYPE