公开(公告)号：US20200005426A1

公开(公告)日：2020-01-02

申请号：US16046785

申请日：2018-07-26

Applicant: QUALCOMM Incorporated

Inventor： Sandeep Nellikatte Srivatsa ,  Anish Kumar ,  Vikash Kumar ,  Ashish Mishra

IPC: G06T1/60 ,  G06T1/20 ,  H04N19/423 ,  H04N21/426 ,  H04N21/443

Abstract: Certain aspects of the present disclosure provide techniques for scalably and efficiently converting linear image data into multi-dimensional image data for multimedia applications. In one example, a method for managing image data includes receiving a line of image data in a linear format via a system bus of width T, wherein the image data's native format is a tile format of H lines per tile; forming H subsets of image data from the line of image data in the linear format; writing the H subsets of image data to a memory comprising BN=H banks of BW=T/BN pixel width, wherein each subset of the H subsets is written to a different bank of the BN banks; and outputting the H subsets of image data in the tile format.

公开(公告)号：US10356440B2

公开(公告)日：2019-07-16

申请号：US14503920

申请日：2014-10-01

Applicant: QUALCOMM Incorporated

Inventor： Ashish Mishra ,  Vikash Kumar

IPC: H04N19/423 ,  H04N19/625 ,  H04N19/44 ,  H04N19/61 ,  H04N19/91 ,  G06F17/16 ,  H04N19/426

Abstract: As the quality and quantity of shared video content increases, video encoding standards and techniques are being developed and improved to reduce bandwidth consumption over telecommunication and other networks. One such technique for compressing videos involves transforming image data into an alternate, encoding-friendly domain (e.g., by a two-dimensional discrete cosine transform). Transform modules may be implemented to perform these transformations, which may occur during both video encoding and decoding processes. Provided are exemplary techniques for improving the efficiency and performance of transform module implementations.

公开(公告)号：US20160100193A1

公开(公告)日：2016-04-07

申请号：US14503920

申请日：2014-10-01

Applicant: QUALCOMM Incorporated

Inventor： Ashish Mishra ,  Vikash Kumar

IPC: H04N19/625 ,  H04N19/423 ,  H04N19/44 ,  H04N19/61 ,  G06F17/16 ,  H04N19/91

CPC classification number: H04N19/625 ,  G06F17/16 ,  H04N19/423 ,  H04N19/427 ,  H04N19/44 ,  H04N19/61 ,  H04N19/91

Abstract: As the quality and quantity of shared video content increases, video encoding standards and techniques are being developed and improved to reduce bandwidth consumption over telecommunication and other networks. One such technique for compressing videos involves transforming image data into an alternate, encoding-friendly domain (e.g., by a two-dimensional discrete cosine transform). Transform modules may be implemented to perform these transformations, which may occur during both video encoding and decoding processes. Provided are exemplary techniques for improving the efficiency and performance of transform module implementations.

Abstract translation: 随着共享视频内容的质量和数量的增加，正在开发和改进视频编码标准和技术，以减少电信和其他网络的带宽消耗。 用于压缩视频的一种这样的技术涉及将图像数据变换成替代的编码友好域（例如，通过二维离散余弦变换）。 可以实现变换模块以执行这些转换，这可能在视频编码和解码过程中发生。 提供了用于提高变换模块实现的效率和性能的示例性技术。

公开(公告)号：US10580112B2

公开(公告)日：2020-03-03

申请号：US16046785

申请日：2018-07-26

Applicant: QUALCOMM Incorporated

Inventor： Sandeep Nellikatte Srivatsa ,  Anish Kumar ,  Vikash Kumar ,  Ashish Mishra

IPC: G09G5/393 ,  G06T1/60 ,  H04N21/426 ,  G06T1/20 ,  H04N21/443 ,  H04N19/423

Abstract: Certain aspects of the present disclosure provide techniques for scalably and efficiently converting linear image data into multi-dimensional image data for multimedia applications. In one example, a method for managing image data includes receiving a line of image data in a linear format via a system bus of width T, wherein the image data's native format is a tile format of H lines per tile; forming H subsets of image data from the line of image data in the linear format; writing the H subsets of image data to a memory comprising BN=H banks of BW=T/BN pixel width, wherein each subset of the H subsets is written to a different bank of the BN banks; and outputting the H subsets of image data in the tile format.