公开(公告)号：US10277904B2

公开(公告)日：2019-04-30

申请号：US14839666

申请日：2015-08-28

Applicant: QUALCOMM Incorporated

Inventor： Shu Lin ,  Mang Li ,  Kai Wang ,  Atul Garg

IPC: H04N19/15 ,  H04N19/152 ,  H04N19/66 ,  H04N19/88 ,  H04N19/423 ,  H04N19/433

Abstract: Video pixel line buffers are widely used for data processing in video codecs. Video data may be packed into buffers configured to store a plurality of words, each word comprising a series of bits. The video data may be associated with two or more channels. In order to reduce realization costs, data blocks from two different channels may be packed from opposite sides of a word in the buffer in opposite directions. In some embodiments, data blocks from two or more physical channels may be mapped to two or more virtual channels, the virtual channels having balanced data block sizes. The data blocks associated with the virtual channels may then be packed to one or more buffers.

公开(公告)号：US20150215617A1

公开(公告)日：2015-07-30

申请号：US14274215

申请日：2014-05-09

Applicant: QUALCOMM Incorporated

Inventor： Athanasios Leontaris ,  Mang Li ,  Sumit Mohan ,  In Suk Chong ,  Hariharan Ganesh Lalgudi

IPC: H04N19/107 ,  H04N19/14 ,  H04N19/192 ,  H04N19/18 ,  H04N19/91 ,  H04N19/82 ,  H04N19/156

CPC classification number: H04N19/82 ,  H04N19/117 ,  H04N19/147 ,  H04N19/176 ,  H04N19/182

Abstract: At least one processor is configured to encode samples of a largest coding unit (LCU) of a picture using a sample adaptive offset (SAO) mode. To encode the samples of the LCU using SAO, the at least one processor is configured to: calculate differences between corresponding reconstructed samples of the LCU and original samples of the LCU, clip a number of bits from each of the differences to form clipped differences, sum the clipped differences to form a sum of differences, clip the sum of differences to form a clipped sum of differences, calculate a number of the reconstructed samples, clip a number of bits from the number of reconstructed samples to form a dipped number of samples, and divide the clipped sum of differences by the clipped number of samples to produce an offset for the LCU.

Abstract translation: 至少一个处理器被配置为使用采样自适应偏移（SAO）模式对图像的最大编码单元（LCU）的采样进行编码。 为了使用SAO对LCU的样本进行编码，所述至少一个处理器被配置为：计算LCU的相应重建样本与LCU的原始样本之间的差异，从每个差异剪切位数以形成裁剪的差异， 将剪切的差值相加以形成差值之和，剪除差分之和以形成限幅差值之和，计算重构样本的数量，从重构样本的数量中剪除一些位数以形成浸渍数量的样本 ，并将限制的差值和除以剪切数量的样本，以产生LCU的偏移量。

公开(公告)号：US09628822B2

公开(公告)日：2017-04-18

申请号：US14274215

申请日：2014-05-09

Applicant: QUALCOMM Incorporated

Inventor： Athanasios Leontaris ,  Mang Li ,  Sumit Mohan ,  In Suk Chong ,  Hariharan Ganesh Lalgudi

IPC: H04N7/26 ,  H04N19/117 ,  H04N19/82 ,  H04N19/176 ,  H04N19/44 ,  H04N19/86 ,  H04N19/14 ,  H04N19/147 ,  H04N19/182

CPC classification number: H04N19/82 ,  H04N19/117 ,  H04N19/147 ,  H04N19/176 ,  H04N19/182

Abstract: At least one processor is configured to encode samples of a largest coding unit (LCU) of a picture using a sample adaptive offset (SAO) mode. To encode the samples of the LCU using SAO, the at least one processor is configured to: calculate differences between corresponding reconstructed samples of the LCU and original samples of the LCU, clip a number of bits from each of the differences to form clipped differences, sum the clipped differences to form a sum of differences, clip the sum of differences to form a clipped sum of differences, calculate a number of the reconstructed samples, clip a number of bits from the number of reconstructed samples to form a dipped number of samples, and divide the clipped sum of differences by the clipped number of samples to produce an offset for the LCU.

公开(公告)号：US20170064308A1

公开(公告)日：2017-03-02

申请号：US14839666

申请日：2015-08-28

Applicant: QUALCOMM Incorporated

Inventor： Shu Lin ,  Mang Li ,  Kai Wang ,  Atul Garg

IPC: H04N19/15 ,  H04N19/66 ,  H04N19/88 ,  H04N19/152

CPC classification number: H04N19/15 ,  H04N19/152 ,  H04N19/423 ,  H04N19/433 ,  H04N19/66 ,  H04N19/88

Abstract: Video pixel line buffers are widely used for data processing in video codecs. Video data may be packed into buffers configured to store a plurality of words, each word comprising a series of bits. The video data may be associated with two or more channels. In order to reduce realization costs, data blocks from two different channels may be packed from opposite sides of a word in the buffer in opposite directions. In some embodiments, data blocks from two or more physical channels may be mapped to two or more virtual channels, the virtual channels having balanced data block sizes. The data blocks associated with the virtual channels may then be packed to one or more buffers.

Abstract translation: 视频像素线缓冲器广泛用于视频编解码器中的数据处理。 视频数据可以被打包到被配置为存储多个字的缓冲器中，每个字包括一系列位。 视频数据可以与两个或更多个信道相关联。 为了降低实现成本，来自两个不同通道的数据块可以从相反方向的缓冲器中的字的相对侧打包。 在一些实施例中，来自两个或多个物理信道的数据块可以被映射到两个或更多个虚拟信道，虚拟信道具有平衡的数据块大小。 然后可以将与虚拟通道相关联的数据块打包到一个或多个缓冲器。

公开(公告)号：US20190174135A1

公开(公告)日：2019-06-06

申请号：US16252231

申请日：2019-01-18

Applicant: QUALCOMM Incorporated

Inventor： Shu Lin ,  Mang Li ,  Kai Wang ,  Atul Garg

IPC: H04N19/15 ,  H04N19/433 ,  H04N19/423 ,  H04N19/152 ,  H04N19/66 ,  H04N19/88

Abstract: Video pixel line buffers are widely used for data processing in video codecs. Video data may be packed into buffers configured to store a plurality of words, each word comprising a series of bits. The video data may be associated with two or more channels. In order to reduce realization costs, data blocks from two different channels may be packed from opposite sides of a word in the buffer in opposite directions. In some embodiments, data blocks from two or more physical channels may be mapped to two or more virtual channels, the virtual channels having balanced data block sizes. The data blocks associated with the virtual channels may then be packed to one or more buffers.