发明名称 VLC coefficient coding for large chroma block
摘要 This disclosure describes techniques for coding transform coefficients for a block of video data. According to these techniques, a video coder (a video encoder or video decoder) determines whether a block of video data is a luma block or a chroma block. If the block of video data is a luma block, the video coder adaptively updates a VLC table index value based on a code number cn and value of a scaling factor. However, if the block of video data is a chroma block, the video coder adaptively updates the VLC table index value based on the code number cn and without using the scaling factor. The video coder uses the updated VLC table index value to select a VLC table of a plurality of VLC tables that are used to encode or decode the block of video data.
申请公布号 US9516316(B2) 申请公布日期 2016.12.06
申请号 US201213535536 申请日期 2012.06.28
申请人 QUALCOMM Incorporated 发明人 Karczewicz Marta;Wang Xianglin;Chien Wei-Jung;Guo Liwei
分类号 H04N19/61;H04N19/91;H04N19/176;H04N19/13;H04N19/60;H04N19/186 主分类号 H04N19/61
代理机构 Shumaker & Sieffert, P.A. 代理人 Shumaker & Sieffert, P.A.
主权项 1. A method of coding luma blocks and chroma blocks of video data, comprising: obtain a plurality of blocks of video data; responsive to determining that a first block of the plurality of blocks of video data is a luma block that is greater than a threshold size, adaptively updating a variable length coding (VLC) table index value used to select a VLC table of a plurality of VLC tables based on a scaling factor and a code number that is based on a position of a last non-zero transform coefficient of the first block of video data in a scan order and whether a magnitude of the last non-zero coefficient of the first block of video data is one or greater than one; responsive to determining that a second block of the plurality of blocks of video data is a luma block that is not greater than the threshold size, adaptively updating the VLC table index value without using the scaling factor and based on a code number that is based on a position of a last non-zero transform coefficient of the second block of video data in the scan order and whether a magnitude of the last non-zero coefficient of the second block of video data is one or greater than one; responsive to determining that a third block of the plurality of blocks of video data is a chroma block of any size, adaptively updating the VLC table index value without using the scaling factor and based on a code number that is based on a position of a last non-zero transform coefficient of the third block of video data in the scan order and whether a magnitude of the last non-zero coefficient of the third block of video data is one or greater than one; and encoding or decoding, based on the updated VLC table index value, a next block of the plurality of blocks of video data.
地址 San Diego CA US