-
1.
公开(公告)号:US20190005396A1
公开(公告)日:2019-01-03
申请号:US15878888
申请日:2018-01-24
申请人: NXP USA, Inc.
CPC分类号: G06N5/045 , G05D1/0088 , G06F16/285 , G06F16/9017 , G06K9/00805 , G06K9/00986 , G06K9/6282 , G06N20/00
摘要: A processing engine for classifying data according to a decision tree having n-nodes is disclosed, wherein each node is represented by a respective test according to which a flag may be set or unset, comprising: a respective test unit and corresponding to each node, having an output flag and being configured to set or unset the respective output flag according to an output of the respective test; a memory configured to hold an n-bit word, each bit corresponding to a one of the respective output flags; and a data-structure configured as a look up table, each entry of the look up table representing a class of the data. Corresponding methods are also disclosed, as are devices and systems incorporating such processing engines.
-
公开(公告)号:US20180357744A1
公开(公告)日:2018-12-13
申请号:US15780052
申请日:2016-12-16
申请人: STC.UNM
CPC分类号: G06T1/20 , G06F17/153 , G06K9/00986 , G06K9/4642 , G06K9/522 , G06T2210/52
摘要: Fast and scalable architectures and methods adaptable to available resources, that (1) compute 2-D convolutions using 1-D convolutions, (2) provide fast transposition and accumulation of results for computing fast cross-correlations or 2-D convolutions, and (3) provide parallel computations using pipelined 1-D convolvers. Additionally, fast and scalable architectures and methods that compute 2-D linear convolutions using Discrete Periodic Radon Transforms (DPRTs) including the use of scalable DPRT, Fast DPRT, and fast 1-D convolutions.
-
公开(公告)号:US20180268235A1
公开(公告)日:2018-09-20
申请号:US15984358
申请日:2018-05-20
申请人: Guobiao ZHANG
发明人: Guobiao ZHANG
CPC分类号: G06K9/00986 , G06F16/51 , G06F16/5838 , G06F21/566 , G06F2221/034 , G10L15/28 , G10L25/51 , H01L27/11206 , H01L27/1128 , H01L27/11551 , H01L27/11578 , H01L27/11582 , H01L27/249
摘要: A preferred image-recognition processor comprises a plurality of storage-processing units (SPU), with each SPU comprising at least a three-dimensional memory (3D-M) array vertically stacked above a pattern-processing circuit. The 3D-M array stores at least a portion of an image model from an image model database. The image data from the input are sent to all SPUs, which perform pattern recognition simultaneously.
-
公开(公告)号:US20180219049A1
公开(公告)日:2018-08-02
申请号:US15505530
申请日:2016-08-17
发明人: Jianfeng Zhu
IPC分类号: H01L27/32 , G09G3/3258 , G09G3/3266 , G06F3/041 , G06F3/0488 , H01L51/52
CPC分类号: H01L27/323 , G06F3/0412 , G06F3/0416 , G06F3/044 , G06F3/0488 , G06F2203/04806 , G06F2203/04808 , G06K9/0004 , G06K9/00986 , G06K9/209 , G06K9/22 , G09G3/3233 , G09G3/3258 , G09G3/3266 , H01L27/3262 , H01L51/5206 , H01L51/5221 , H04M2250/22
摘要: A driving and scanning circuit, a display screen and a mobile terminal, comprise driving and scanning units which are distributed in an array, each comprising an AMOLED driving unit for driving an OLED to emit light and a fingerprint scanning unit for forming pixel capacitances, wherein the AMOLED driving unit comprises a first thin film transistor, a second thin film transistor, a third thin film transistor and an OLED, the first thin film transistor is connected with a driving voltage and a first switching voltage, the second thin film transistor and the third thin film transistor are respectively connected with an anode terminal and a cathode terminal, and the OLED is positioned between the cathode terminal and the second thin film transistor; the cathode terminal also comprises a fourth thin film transistor connected with a second driving voltage and a second switching voltage.
-
公开(公告)号:US20180204307A1
公开(公告)日:2018-07-19
申请号:US15874611
申请日:2018-01-18
申请人: NVIDIA Corporation
发明人: Christoph H. Schied , Marco Salvi , Anton S. Kaplanyan , Aaron Eliot Lefohn , John Matthew Burgess , Anjul Patney , Christopher Ryan Wyman
CPC分类号: G06T5/20 , G06K9/00986 , G06K9/6273 , G06K9/66 , G06T1/20 , G06T5/002 , G06T11/60 , G06T2207/10016 , G06T2207/10024 , G06T2207/10028 , G06T2207/20081 , G06T2207/20084 , G06T2207/20182
摘要: A method, computer readable medium, and system are disclosed for performing spatiotemporal filtering. The method includes the steps of applying, utilizing a processor, a temporal filter of a filtering pipeline to a current image frame, using a temporal reprojection, to obtain a color and auxiliary information for each pixel within the current image frame, providing the auxiliary information for each pixel within the current image frame to one or more subsequent filters of the filtering pipeline, and creating a reconstructed image for the current image frame, utilizing the one or more subsequent filters of the filtering pipeline.
-
公开(公告)号:US10019788B1
公开(公告)日:2018-07-10
申请号:US15432834
申请日:2017-02-14
申请人: Cogniac, Corp.
发明人: William S Kish , Huayan Wang , Sandip C. Patel , Yui Ming Tsang
CPC分类号: G06T7/0002 , G06K9/00369 , G06K9/00778 , G06K9/00979 , G06K9/00986 , G06K9/4604 , G06K9/4628 , G06T7/001 , G06T2207/10024 , G06T2207/20081 , G06T2207/20084 , G06T2207/30108
摘要: A computer system may train and use a machine-learning model to quantitatively analyze an image. In particular, the computer system may generate the machine-learning model based on a set of reference images that include content with instances of a quantitative feature attribute and one or more feedback metrics that specify locations of the instances of the quantitative feature attribute in the reference images and numerical values associated with the instances of the quantitative feature attribute. Then, after receiving the image from an electronic device, the computer system may analyze the image using the machine-learning model to perform measurements of one or more additional instances of the quantitative feature attribute in the image. Moreover, the computer system may provide a measurement result for the image, the measurement result including a second numerical value associated with the one or more additional instances of the quantitative feature attribute in the image.
-
公开(公告)号:US20180189537A1
公开(公告)日:2018-07-05
申请号:US15397709
申请日:2017-01-03
发明人: Jia-Ming HE , Yaw-Guang CHANG
CPC分类号: G06K9/0002 , G06K9/0008 , G06K9/00986 , H03K17/962 , H03K17/9622 , H03K2217/960725 , H03K2217/960735
摘要: The fingerprint sensing circuit including sensing electrodes, switches, a sensing line and flip flops. First terminals of the switches are respectively coupled to the sensing electrodes. The sensing line is coupled to second terminals of the switches. Output terminals of the flip flops are respectively coupled to control terminals of the switches. An input terminal of part of the flip flops is coupled to the output terminal of a previous flip flop.
-
公开(公告)号:US10013628B2
公开(公告)日:2018-07-03
申请号:US14662488
申请日:2015-03-19
发明人: Masami Kato , Hirotaka Hachiya
CPC分类号: G06K9/4628 , G06K9/00986
摘要: A plurality of random number sequences are generated using a plurality of random number seeds stored in a memory, and a plurality of base vectors are generated based on the plurality of random number sequences. A transformation source vector is transformed into a lower-dimensional vector by performing random projection for the transformation source vector using the plurality of generated base vectors.
-
公开(公告)号:US20180137416A1
公开(公告)日:2018-05-17
申请号:US15871660
申请日:2018-01-15
发明人: David R. Brown , Harold B Noyes
CPC分类号: G06N3/08 , G06K9/00986 , G06N3/063
摘要: A device includes a match element that includes a first data input configured to receive a first result, wherein the first result is of an analysis performed on at least a portion of a data stream by an element of a state machine. The match element also includes a second data input configured to receive a second result, wherein the second result is of an analysis performed on at least a portion of the data stream by another element of the state machine. The match element further includes an output configured to selectively provide the first result or the second result.
-
公开(公告)号:US09971959B2
公开(公告)日:2018-05-15
申请号:US14029640
申请日:2013-09-17
申请人: NVIDIA CORPORATION
CPC分类号: G06K9/6285 , G06K9/00986 , G06K9/6256 , G06K9/6282
摘要: In one embodiment of the present invention, a graphics processing unit (GPU) is configured to detect an object in an image using a random forest classifier that includes multiple, identically structured decision trees. Notably, the application of each of the decision trees is independent of the application of the other decision trees. In operation, the GPU partitions the image into subsets of pixels, and associates an execution thread with each of the pixels in the subset of pixels. The GPU then causes each of the execution threads to apply the random forest classifier to the associated pixel, thereby determining a likelihood that the pixel corresponds to the object. Advantageously, such a distributed approach to object detection more fully leverages the parallel architecture of the parallel processing unit (PPU) than conventional approaches. In particular, the PPU performs object detection more efficiently using the random forest classifier than using a cascaded classifier.
-
-
-
-
-
-
-
-
-