Method and apparatus for implementing a nearest neighbor search on a graphics processing unit (GPU)

    公开(公告)号:US10032244B2

    公开(公告)日:2018-07-24

    申请号:US14581490

    申请日:2014-12-23

    Abstract: An apparatus and method are described for implementing a nearest neighbor search on a graphics processing unit. For example, one embodiment of an apparatus comprises: a host processor; and a graphics processor unit (GPU) to execute a nearest neighbor (NN) tree build operation and a NN tree search operation in response to signals from the host processor; the NN tree build operation comprising dividing a first set of data points to define each tree node to form a plurality of buckets, each bucket comprising at least one of the first set of data points, the number of buckets to be limited based on a threshold value, resulting in at least some buckets with multiple data points; and the NN tree search operation comprising traversing the tree by comparing each of a second set of input data points to data points in the tree to identify a bucket and then comparing one of the second data points to data points in the bucket to identify a nearest neighbor data point in the bucket.

    Method and apparatus for natural hand visualization

    公开(公告)号:US10559119B2

    公开(公告)日:2020-02-11

    申请号:US15692856

    申请日:2017-08-31

    Abstract: An apparatus and method for natural hand processing for virtual reality. For example, one embodiment of an apparatus comprises: graphics processing circuitry to render left and right scenes responsive to a virtual reality application, the left and right scenes to be displayed on left and right displays, respectively, of a virtual reality device; and a natural hand processing pipeline to project a representation of a user's hands into the left and right scenes to allow the user to interact with one or more objects in the scenes, the natural hand processing pipeline comprising: a depth/mask data smoothing and reconstruction module to receive noisy depth and mask data associated with images of the user's hands and to enhance the noisy depth data and mask data to generate enhanced depth and mask data; a hand mesh generator to generate a hand mesh using the enhanced depth and mask data; and a mesh data projector to project the hand mesh into the left and right scenes.

    Systems, methods, and apparatus to improve computing system utilization

    公开(公告)号:US10956789B2

    公开(公告)日:2021-03-23

    申请号:US16368562

    申请日:2019-03-28

    Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to improve computing system utilization. An example system includes a visual recognition controller to determine whether an infrared image of the device corresponds to a first utilization state of the device based on a machine-learning model, and generate a report including the first utilization state, and at least one processor to execute one or more actions to adjust operation of the device from the first utilization state to a second utilization state based on the report obtained from the visual recognition controller, the second utilization state corresponding to reduced utilization compared to the first utilization state.

    Connected component labeling in graphics processors
    4.
    发明授权
    Connected component labeling in graphics processors 有权
    在图形处理器中连接的组件标签

    公开(公告)号:US09558560B2

    公开(公告)日:2017-01-31

    申请号:US14210585

    申请日:2014-03-14

    Abstract: Systems and methods may provide for obtaining data associated with an image and using a plurality of threads in a graphics processor to conduct a single instruction multiple data (SIMD) scan of the data. Additionally, systems and methods may provide for generating a plurality of connection tables corresponding to the plurality of threads based on the SIMD scan. In one example, a plurality of threads in the graphics processor are used to conduct a single phase merge of the plurality of connection tables onto a global connected components labeling (CCL) table for the image.

    Abstract translation: 系统和方法可以提供用于获得与图像相关联的数据并且使用图形处理器中的多个线程来进行数据的单指令多数据(SIMD)扫描。 此外,系统和方法可以提供用于基于SIMD扫描生成与多个线程相对应的多个连接表。 在一个示例中,图形处理器中的多个线程用于将多个连接表的单相合并进行到用于图像的全局连接分量标签(CCL)表。

Patent Agency Ranking