Inventor · disambiguated record
Cagatay Dikici
Also filed as: DIKICI CAGATAY
16 granted patents·46 pending applications·7 citations·filing 2018–2025
87Inventor score
Files withIMAGINATION TECH LTD62
Top patents by PatentIndex Score
62 records- 0191US11188817B2Methods and systems for converting weights of a deep neural network from a first number format to a second number formatIMAGINATION TECH LTD·Filed 2020·Granted Nov 30, 2021·3 cites·20 claims
- 0290US11886536B2Methods and systems for implementing a convolution transpose layer of a neural networkIMAGINATION TECH LTD·Filed 2023·Granted Jan 30, 2024·1 cites·20 claims
- 0386US11556613B2Methods and systems for implementing a convolution transpose layer of a neural networkIMAGINATION TECH LTD·Filed 2020·Granted Jan 17, 2023·2 cites·17 claims
- 0482US2025068693A1Methods and systems for implementing a convolution transpose layer of a neural networkIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 0581US12174910B2Methods and systems for implementing a convolution transpose layer of a neural networkIMAGINATION TECH LTD·Filed 2024·Granted Dec 24, 2024·0 cites·20 claims
- 0680US2024320778A1Methods and systems for performing a sparse submanifold convolution on a gpuIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 0780US2025068883A1Hierarchical mantissa bit length selection for hardware implementation of deep neural networkIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 0880US2024320297A1Methods and systems for performing a standard convolution on a gpuIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 0980US2024320298A1Methods and systems for performing a sparse submanifold convolution using an nnaIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 1080US2024320779A1Methods and systems for performing a sparse submanifold deconvolution on a gpuIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 1180US2024320299A1Methods and systems for performing a standard deconvolution on a gpuIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 1279US2025315961A1Estimating motion of objects in a set of frames by matching object-instances in the frames based on feature vectors and masksIMAGINATION TECH LTD·Filed 2025·Application pending·0 cites
- 1378US2024394525A1Histogram-based per-layer data format selection for hardware implementation of deep neural networkIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 1477US12347119B2Training a machine learning algorithm to perform motion estimation of objects in a set of framesIMAGINATION TECH LTD·Filed 2024·Granted Jul 1, 2025·0 cites·20 claims
- 1576US2024412056A1Hardware implementation of a deep neural network with variable output data formatIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 1675US12056600B2Histogram-based per-layer data format selection for hardware implementation of deep neural networkIMAGINATION TECH LTD·Filed 2023·Granted Aug 6, 2024·0 cites·20 claims
- 1772US2024169025A1Implementing a scatter function on a neural network acceleratorIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 1872US2024160692A1Implementing a scatter function on a neural network acceleratorIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 1972US2024169024A1Implementing a scatter function on a neural network acceleratorIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 2072US2024320480A1Compressing a neural networkIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 2172US2024354574A1Compressing a neural networkIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 2270US2024354560A1Hardware implementation of a neural networkIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 2368US2024135139A1Implementing Traditional Computer Vision Algorithms as Neural NetworksIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 2467US11593626B2Histogram-based per-layer data format selection for hardware implementation of deep neural networkIMAGINATION TECH LTD·Filed 2018·Granted Feb 28, 2023·1 cites·19 claims
- 2565US2025181564A1Indexing Elements in a Source ArrayIMAGINATION TECH LTD·Filed 2025·Application pending·0 cites
- 2664US12073567B2Analysing objects in a set of framesIMAGINATION TECH LTD·Filed 2021·Granted Aug 27, 2024·0 cites·18 claims
- 2764US2025322222A1Attention-based neural networksIMAGINATION TECH LTD·Filed 2025·Application pending·0 cites
- 2864US2025315645A1Graph convolutional networksIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 2964US2025322221A1Attention-based neural networksIMAGINATION TECH LTD·Filed 2025·Application pending·0 cites
- 3062US12229105B2Indexing elements in a source arrayIMAGINATION TECH LTD·Filed 2021·Granted Feb 18, 2025·0 cites·17 claims
- 3162US2024127044A1Hardware implementation of an attention-based neural networkIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 3262US2025021805A1Compressing a neural networkIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 3362US2025181921A1Training a neural networkIMAGINATION TECH LTD·Filed 2025·Application pending·0 cites
- 3462US2024127049A1Hardware implementation of an attention-based neural networkIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 3561US12260336B2Training a neural networkIMAGINATION TECH LTD·Filed 2021·Granted Mar 25, 2025·0 cites·20 claims
- 3660US12488253B2Neural network comprising matrix multiplicationIMAGINATION TECH LTD·Filed 2022·Granted Dec 2, 2025·0 cites·20 claims
- 3760US2025148264A1Hardware implementation of windowed operations in three or more dimensionsIMAGINATION TECH LTD·Filed 2025·Application pending·0 cites
- 3859US12198034B2Hardware implementation of windowed operations in three or more dimensionsIMAGINATION TECH LTD·Filed 2021·Granted Jan 14, 2025·0 cites·16 claims
- 3959US2024119254A1Mapping neural networks to hardwareIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 4058US12061972B2Hardware implementation of a neural networkIMAGINATION TECH LTD·Filed 2020·Granted Aug 13, 2024·0 cites·20 claims
- 4157US2025322203A1Compressing a graph attention networkIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 4257US2024143986A1Methods and systems for executing a neural network on a neural network acceleratorIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 4356US2024232600A1Variable input shapes at runtimeIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 4456US2024346108A1System and method of performing convolution efficiently adapting winograd algorithmIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 4555US12175349B2Hierarchical mantissa bit length selection for hardware implementation of deep neural networkIMAGINATION TECH LTD·Filed 2018·Granted Dec 24, 2024·0 cites·20 claims
- 4655US2024346107A1Method and data processing system for resampling a set of samplesIMAGINATION TECH LTD·Filed 2024·Application pending·0 cites
- 4754US11636306B2Implementing traditional computer vision algorithms as neural networksIMAGINATION TECH LTD·Filed 2019·Granted Apr 25, 2023·0 cites·17 claims
- 4854US2024160908A1Methods and systems for online selection of number formats for network parameters of a neural networkIMAGINATION TECH LTD·Filed 2023·Application pending·0 cites
- 4952US12165045B2Hardware implementation of a deep neural network with variable output data formatIMAGINATION TECH LTD·Filed 2018·Granted Dec 10, 2024·0 cites·20 claims
- 5051US2022253506A1Implementing dilated convolution in hardwareIMAGINATION TECH LTD·Filed 2022·Application pending·0 cites
Showing the top 50 of 62 patent records by PatentIndex Score.
Join the waitlist — get patent alerts
Get an alert when Cagatay Dikici files or is granted a new patent.
We store only your email — no account needed. See our privacy policy.
Identity basis: PatentsView inventor disambiguation (2025Q4-odp release). How scoring works →