Bandwidth efficient method for generating an alpha hint buffer

    公开(公告)号:US10121222B2

    公开(公告)日:2018-11-06

    申请号:US15242478

    申请日:2016-08-19

    申请人: MediaTek Inc.

    IPC分类号: G06T1/60 G06T11/60

    摘要: A graphics processing unit (GPU) renders graphical objects into a group of pixels and stores the pixels in an on-chip buffer on the same chip as the GPU. Each pixel has an alpha value that indicates transparency of the pixel. The GPU reads the alpha value of each pixel from the on-chip buffer. According to alpha values of the group of pixels, the GPU generates an alpha hint in the system memory for the group of pixels. The alpha hint represents an aggregate of the alpha values of the group of pixels. The GPU then stores the group of pixels into a frame buffer in the system memory.

    Graphic processing circuit with binning rendering and pre-depth processing method thereof
    8.
    发明授权
    Graphic processing circuit with binning rendering and pre-depth processing method thereof 有权
    具有binning渲染的图形处理电路及其深度处理方法

    公开(公告)号:US09361697B1

    公开(公告)日:2016-06-07

    申请号:US14582134

    申请日:2014-12-23

    申请人: MediaTek Inc.

    CPC分类号: G06T15/005 G06T15/40

    摘要: A graphic processing circuit with binning rendering and associated pre-depth processing method is provided. Firstly, a first depth data of a first primitive corresponding to a specified tile is received. Then, the pre-depth data corresponding to the specified tile is read from a pre-Z buffer. If the first depth data is not larger than the pre-depth data and the first primitive is an opaque primitive, the pre-depth data is updated with the first depth data. If the first depth data is not larger than the pre-depth data and the first primitive is a translucent primitive, an uncertainty ordering range is defined according to the first depth data and the pre-depth data, and the pre-depth data is updated with the uncertainty ordering range.

    摘要翻译: 提供了具有binning渲染和相关的预深度处理方法的图形处理电路。 首先,接收对应于指定瓦片的第一原语的第一深度数据。 然后,从预Z缓冲器读取对应于指定瓦片的预深度数据。 如果第一深度数据不大于预深度数据,并且第一原语是不透明原语,则用第一深度数据更新预深度数据。 如果第一深度数据不大于预深度数据,并且第一图元是半透明图元,则根据第一深度数据和预深度数据来定义不确定性排序范围,并且更新预深度数据 具有不确定性排序范围。

    AI frame engine for mobile edge
    10.
    发明授权

    公开(公告)号:US11580621B2

    公开(公告)日:2023-02-14

    申请号:US17113397

    申请日:2020-12-07

    申请人: MEDIATEK INC.

    摘要: Aspects of the disclosure provide a device for processing frames with aliasing artifacts. For example, the device can include a motion estimation circuit, a warping circuit coupled to the motion estimation circuit, and a temporal decision circuit coupled to the warping circuit. The motion estimation circuit can estimate a motion value between a current frame and a previous frame. The warping circuit can warp the previous frame based on the motion value such that the warped previous frame is aligned with the current frame and determine whether the current frame and the warped previous frame are consistent. The temporal decision circuit can generate an output frame, the output frame including either the current frame and the warped previous frame when the current frame and the warped previous frame are consistent, or the current frame when the current frame and the warped previous frame are not consistent.