-
公开(公告)号：US20230012553A1
公开(公告)日：2023-01-19
申请号：US17850319
申请日：2022-06-27
Applicant: Imagination Technologies Limited
Inventor： Aria Ahmadi , Muhammad Asad , Cagatay Dikici , Elia Condorelli
Abstract: A mechanism for processing, on a hardware accelerator comprising fixed-function circuitry, data according to a neural network process that comprises a neural network with an associated argmax or argmin function. The argmax or argmin function is mapped to a set of elementary neural network operations available to the fixed-function circuitry. The neural network process is then executed using the fixed-function circuitry. The data processed using the neural network process comprises image and/or audio data.
-
公开(公告)号：US20220253716A1
公开(公告)日：2022-08-11
申请号：US17568325
申请日：2022-01-04
Applicant: Imagination Technologies Limited
Inventor： Biswarup Choudhury , Aria Ahmadi , James Imber , Cagatay Dikici , Timothy Atherton
Abstract: A method and data processing system implement a neural network containing at least one matrix multiplication operation. The matrix multiplication operation is mapped to a graph of neural network operations including at least one transformation and at least one convolution. The at least one convolution is implemented in fixed-function hardware of a neural network accelerator.
-
公开(公告)号：US20220253506A1
公开(公告)日：2022-08-11
申请号：US17583411
申请日：2022-01-25
Applicant: Imagination Technologies Limited
Inventor： Aria Ahmadi , Cagatay Dikici , Clement Charnay
Abstract: A method and data processing system implement dilated convolution operations in hardware. Embodiments provide various ways to implement a dilated convolution based on a number of constituent convolutions, by either splitting the kernel to construct a set of constituent convolutions with smaller kernels, or dividing the input data into multiple parts and applying a convolution to each part separately. The constituent convolutions are evaluated in hardware and their results are combined to produce the result of the dilated convolution.
-
公开(公告)号：US20210272295A1
公开(公告)日：2021-09-02
申请号：US17187831
申请日：2021-02-28
Applicant: Imagination Technologies Limited
Inventor： Aria Ahmadi , David Walton , Cagatay Dikici
Abstract: A method of analysing objects in a first frame and a second frame is disclosed. The method includes segmenting the frames, and matching at least one object in the first frame with a corresponding object in the second frame. The method optionally includes estimating the motion of the at least one matched object between the frames. Also disclosed is a method of generating a training dataset suitable for training machine learning algorithms to estimate the motion of objects. Also provided are processing systems configured to carry out these methods.
-
公开(公告)号：US20250148264A1
公开(公告)日：2025-05-08
申请号：US19018414
申请日：2025-01-13
Applicant: Imagination Technologies Limited
Inventor： Ivaxi Sheth , Aria Ahmadi , James Imber , Cagatay Dikici
Abstract: A windowed operation is implemented in at least three traversed dimensions. The windowed operation applies a window having at least three dimensions to data having at least three traversed dimensions, with shifts of the window in all three traversed dimensions. Two dimensions of the at least three traversed dimensions are selected, and the windowed operation is mapped to a plurality of constituent 2-D windowed operations in the selected two dimensions, the 2-D windowed operations applying a slice of the window to a slice of the data, with shifts of the slice of the window in only two dimensions. Each of the plurality of 2-D windowed operations is implemented by at least one hardware accelerator, each 2-D windowed operation producing a respective partial result, and the partial results are assembled to produce the result of the windowed operation.
-
Patent Agency Ranking
公开(公告)号：US12229105B2
公开(公告)日：2025-02-18
申请号：US17321091
申请日：2021-05-14
Applicant: Imagination Technologies Limited
Inventor： Aria Ahmadi , Cagatay Dikici
Abstract: A hardware-implemented method of indexing data elements in a source array is provided. The method comprises generating a number of shifted copy arrays; receiving indices for indexing the source array; and retrieving one or more data elements from the shifted copy arrays, according to the received indices. Also disclosed is a related processing system comprising a memory and hardware for indexing data elements in a source array in the memory.
-
公开(公告)号：US20240346107A1
公开(公告)日：2024-10-17
申请号：US18617810
申请日：2024-03-27
Applicant: Imagination Technologies Limited
Inventor： Aria Ahmadi , Cagatay Dikici
CPC classification number: G06F17/15 , G06F9/5027
Abstract: A method and data processing system for resampling a first set of samples using a neural network accelerator. The first set of samples is arranged in a tensor extending in at least a first dimension defined in a first coordinate system. A set of resampling parameters is determined, having a first resampling factor a_1/b_1 for a first dimension, and a first offset d_1 for the first dimension. At least a first number of kernels is obtained, and the first set of samples is resampled to produce a second set of samples, based on the first resampling factor and the first offset.
-
公开(公告)号：US12073567B2
公开(公告)日：2024-08-27
申请号：US17187831
申请日：2021-02-28
Applicant: Imagination Technologies Limited
Inventor： Aria Ahmadi , David Walton , Cagatay Dikici
CPC classification number: G06T7/246 , G06F18/22 , G06F18/23 , G06T7/207 , G06T7/248 , G06T2207/20081
Abstract: A method of analysing objects in a first frame and a second frame is disclosed. The method includes segmenting the frames, and matching at least one object in the first frame with a corresponding object in the second frame. The method optionally includes estimating the motion of the at least one matched object between the frames. Also disclosed is a method of generating a training dataset suitable for training machine learning algorithms to estimate the motion of objects. Also provided are processing systems configured to carry out these methods.
-
公开(公告)号：US20230019151A1
公开(公告)日：2023-01-19
申请号：US17850517
申请日：2022-06-27
Applicant: Imagination Technologies Limited
Inventor： Aria Ahmadi , Cagatay Dikici , Muhammad Asad , Elia Condorelli
Abstract: A mechanism for processing, on a hardware accelerator comprising fixed-function circuitry, data according to a neural network process that includes a pooling, unpooling or backward pooling and/or binary argmax/argmin function. The function is mapped to a set of elementary neural network operations available to the fixed-function circuitry. The neural network process is then executed using the fixed-function circuitry. The data processed using the neural network process comprises image and/or audio data.
-
公开(公告)号：US20220391172A1
公开(公告)日：2022-12-08
申请号：US17681557
申请日：2022-02-25
Applicant: Imagination Technologies Limited
Inventor： James Imber , Biswarup Choudhury , Cagatay Dikici , Timothy Atherton , Aria Ahmadi
Abstract: Methods for implementing an exponential operation, and a softmax neural network layer, in neural network accelerator hardware, and a data processing system for implementing the exponential operation and a data processing system for implementing the softmax layer. The exponential operation or softmax layer is mapped to a plurality of elementary neural network operations, and the neural network accelerator hardware evaluates these operations, to produce the result of the operation or layer respectively.
-
-
-
-
-
-
-
-
-
Search Results
20
records
(took 0.018 seconds)
