SINGLE PASS ANTI-RINGING CLAMPING ENABLED IMAGE PROCESSING

    公开(公告)号:US20240331079A1

    公开(公告)日:2024-10-03

    申请号:US18194324

    申请日:2023-03-31

    CPC classification number: G06T1/20 G06T5/20 G06T5/77 G06T7/40 G06V10/54

    Abstract: This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for single pass anti-ringing clamping enabled image processing. A graphics processor may perform a filtering operation on a set of texture samples. The graphics processor may select, during a single sampling operation, a minimum value and a maximum value associated with the set of texture samples during the performance of the filtering operation on the set of texture samples. The graphics processor may adjust, during the single sampling operation, a value of a filtered texture sample associated with the set of texture samples based on the minimum value and the maximum value. The graphics processor may output an indication of the adjusted value of the filtered texture sample.

    WORKLOAD PACKING IN GRAPHICS TEXTURE PIPELINE

    公开(公告)号:US20240212257A1

    公开(公告)日:2024-06-27

    申请号:US18145804

    申请日:2022-12-22

    CPC classification number: G06T15/04 G06T1/20 G06T15/005

    Abstract: This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for workload packing in a graphics texture pipeline. A graphics processor may combine a set of samples into one or more hardware transactions, where the set of samples is associated with at least one of a first type of texture filtering or a first type of shader requested texture format component and the one or more hardware transactions are associated with at least one of a second type of texture filtering or a second type of shader requested texture format component. The graphics processor may process, in a texture pipeline at the graphics processor, the one or more hardware transactions including the set of samples. The graphics processor may output an indication of the processed one or more hardware transactions including the set of samples.

    METHODS AND APPARATUS FOR TENSOR OBJECT SUPPORT IN MACHINE LEARNING WORKLOADS

    公开(公告)号:US20220253969A1

    公开(公告)日:2022-08-11

    申请号:US17173643

    申请日:2021-02-11

    Abstract: The present disclosure relates to methods and devices for graphics processing including an apparatus, e.g., a GPU. The apparatus may modify at least one texture memory object to support a data structure for one or more tensor objects. The apparatus may also determine one or more supported memory layouts for the one or more tensor objects based on the modified at least one texture memory object. Additionally, the apparatus may access data associated with the one or more tensor objects based on the one or more supported memory layouts, the data for each of the one or more tensor objects corresponding to at least one data instruction. The apparatus may also execute the at least one data instruction based on the accessed data associated with the one or more tensor objects.

    METADATA UPDATING
    5.
    发明公开
    METADATA UPDATING 审中-公开

    公开(公告)号:US20240296153A1

    公开(公告)日:2024-09-05

    申请号:US18177390

    申请日:2023-03-02

    CPC classification number: G06F16/23 G06F16/2282

    Abstract: Aspects of the disclosure are directed to metadata updating. In accordance with one aspect, an apparatus includes an external memory unit configured for storing an original descriptor tag; a descriptor loading block coupled to the external memory, the descriptor loading block configured to fetch the original descriptor tag from the external memory for storage in an internal cache memory and further configured to compare the original descriptor tag stored in the internal cache memory to each of a plurality of original base values; and a remap table database coupled to the descriptor loading block, the remap table database configured to store the plurality of original base values, a plurality of updated base values and a plurality of updated miscellaneous base values.

    BLOCK MATCHING WINDOW SEARCH INSTRUCTION USING A GRAPHICS PROCESSING UNIT (GPU) TEXTURE PIPELINE

    公开(公告)号:US20240242417A1

    公开(公告)日:2024-07-18

    申请号:US18155679

    申请日:2023-01-17

    CPC classification number: G06T15/005 G06F7/57 G06T2210/52

    Abstract: Aspects presented herein relate to methods and devices for graphics processing including an apparatus, e.g., a graphics processing unit (GPU). The apparatus may obtain a first indication of a first image including a set of first image sections with a plurality of first subsections and a second indication of a second image including a set of second image sections with a plurality of second subsections. The apparatus may also compare one first image section with one second image section. Further, the apparatus may calculate a magnitude of difference between one first subsection and each of the multiple second subsections. The apparatus may also output a third indication of at least one of: (1) a lowest magnitude of difference between the one first subsection and each of the multiple second subsections or (2) a set of coordinates for a second subsection that corresponds to the lowest magnitude of difference.

    METHODS AND APPARATUS FOR MAPPING SOURCE LOCATION FOR INPUT DATA TO A GRAPHICS PROCESSING UNIT

    公开(公告)号:US20220036498A1

    公开(公告)日:2022-02-03

    申请号:US16984024

    申请日:2020-08-03

    Abstract: The present disclosure relates to methods and apparatus for mapping a source location of input data for processing by a graphics processing unit. The apparatus can configure a processing element of the graphics processing unit with a predefined rule for decoding a data source parameter for executing a task by the graphics processing unit. Moreover, the apparatus can store the parameter in local storage of the processing element and configure the processing element to decode the parameter according to the at least one predefined rule to determine a source location of the input data and at least one relationship between invocations of the task. The apparatus can also load, to the local storage of the processing element, the input data from a plurality of memory addresses of the source location determined by the parameter. A one logic unit can then execute the task on the loaded input data.

Patent Agency Ranking