-
公开(公告)号：EP4231203A1
公开(公告)日：2023-08-23
申请号：EP23150415.0
申请日：2023-01-05
Inventor： KIM, Jongseok , JEONG, Yeonwoo , SONG, Hyun Oh , SON, Changyong
IPC: G06N3/0495 , G06N3/082
Abstract: A method and apparatus with pruning are disclosed. A method is performed by an apparatus including a processor, the method includes determining weight importance of a trained neural network, receiving a constraint condition related to an operation resource, and determining, in accordance with the constraint condition, a pruning mask for maximizing the weight importance of the trained neural network.
-
公开(公告)号：EP3839832A1
公开(公告)日：2021-06-23
申请号：EP20193180.5
申请日：2020-08-27
Applicant: Samsung Electronics Co., Ltd.
Inventor： SON, Jinwoo , SON, Changyong , YOO, Jaehyoung , LEE, Seohyung , JUNG, Sangil , CHOI, Changin
Abstract: A processor-implemented neural network method includes: generating a first output line of an output feature map by performing a convolution operation between a first input line group of an input feature map and weight kernels; generating a first output of an operation block including the convolution operation based on the first output line; and storing the first output in a memory in which the input feature map is stored by overwriting the first output to a memory space of the memory.
-
公开(公告)号：EP3671575A2
公开(公告)日：2020-06-24
申请号：EP19204904.7
申请日：2019-10-23
Applicant: Samsung Electronics Co., Ltd.
Inventor： LEE, Seoh Yung , KWAK, Youngjun , SON, Changyong , SON, Jinwoo , JUNG, San Gil , CHOI, Chang Kyu , HAN, Jaejoon
Abstract: A neural network processing method and apparatus based on nested bit representation is provided. The processing method includes obtaining first weights for a first layer of a source model of a first layer of a neural network, determining a bit-width for the first layer of the neural network, obtaining second weights for the first layer of the neural network by extracting at least one bit corresponding to the determined bit-width from each of the first weights for the first layer of a source model corresponding to the first layer of the neural network, and processing input data of the first layer of the neural network by executing the first layer of the neural network based on the obtained second weights.
-
公开(公告)号：EP3499412A1
公开(公告)日：2019-06-19
申请号：EP18210386.1
申请日：2018-12-05
Applicant: Samsung Electronics Co., Ltd.
Inventor： HAN, Seungju , Park, SungUn , HAN, Jaejoon , Son, Jinwoo , SON, Changyong , Ko, Minsu , Kim, Jihye , CHOI, Chang Kyu
IPC: G06K9/00
Abstract: An object recognition apparatus and method are provided. The apparatus comprises a processor configured to verify a target image using an object model and based on reference intermediate data extracted by a partial layer of the object model as used in an object recognition of an input image, in response to a failure of a verification of the input image after a success of the object recognition of the input image, and perform an additional verification of the target image in response to the target image being verified in the verifying of the target image.
-
公开(公告)号：EP3382603A1
公开(公告)日：2018-10-03
申请号：EP18164571.4
申请日：2018-03-28
Applicant: Samsung Electronics Co., Ltd.
Inventor： SON, Changyong , KIM, Deoksang , KO, Minsu , SON, Jinwoo , HAN, Seungju , CHOI, Chang Kyu , HAN, Jae-Joon
CPC classification number: G06K9/00288 , G06F21/32 , G06K9/00228 , G06K9/00268 , G06K9/00281 , G06K9/22 , G06K9/3233 , G06K9/6202 , G06K9/6215 , G06K9/6256 , G06K9/629 , G06K9/6292 , G06N3/08 , G06Q20/40145 , G06T3/4038
Abstract: A face verification method and apparatus is disclosed. The face verification method includes selecting a current verification mode, from among plural verification modes, to be implemented for the verifying of the face, determining one or more recognizers, from among plural recognizers, based on the selected current verification mode, extracting feature information from information of the face using at least one of the determined one or more recognizers, and indicating whether a verification is successful based on the extracted feature information.
-
Patent Agency Ranking
公开(公告)号：EP3319015A1
公开(公告)日：2018-05-09
申请号：EP17197450.4
申请日：2017-10-20
Applicant: Samsung Electronics Co., Ltd.
Inventor： SON, Jinwoo , SON, Changyong , CHOI, Chang Kyu , HAN, Jaejoon
CPC classification number: G06K9/00979 , G06K9/6256 , G06N3/04 , G06N3/0454 , G06N3/063 , G06N3/08
Abstract: A convolutional neural network (CNN) processing method and apparatus is disclosed. The apparatus may select, based on at least one of a characteristic of a kernel of a convolution layer or a characteristic of an input of the convolution layer, one operation mode from a first operation mode reusing the kernel and a second operation mode reusing the input, and perform a convolution operation based on the selected operation mode.
Abstract translation: 公开了一种卷积神经网络（CNN）处理方法和装置。 该设备可以基于卷积层的内核的特性或卷积层的输入的特性中的至少一个来选择来自重用内核的第一操作模式和重用该输入的第二操作模式的一个操作模式 并基于所选择的操作模式执行卷积操作。
-
公开(公告)号：EP4064216A1
公开(公告)日：2022-09-28
申请号：EP21207318.3
申请日：2021-11-09
Applicant: Samsung Electronics Co., Ltd.
Inventor： SONG, Ju Hwan , LEE, Hyunjeong , PARK, Changbeom , SON, Changyong , YOO, Byung In
Abstract: A processor-implemented object tracking method includes: setting a suppressed region in a template image based on a shape of a target box of the template image; refining a template feature map of the template image by suppressing an influence of feature data corresponding to the suppressed region in the template feature map; and tracking an object by determining a bounding box corresponding to the target box in a search image based on the refined template feature map.
-
公开(公告)号：EP3859660A1
公开(公告)日：2021-08-04
申请号：EP20214980.3
申请日：2020-12-17
Applicant: Samsung Electronics Co., Ltd.
Inventor： JUNG, Sangil , SON, Changyong , SON, Jinwoo , YOO, Jaehyoung , CHOI, Changin , HAN, Jaejoon
IPC: G06T5/20
Abstract: Disclosed are an image data processing method and a sensor device performing the same. The sensor device includes an image sensor configured to acquire image data, an image buffer configured to store the image data, and an image processor configured to generate image-processed data by applying a filter corresponding to a storage pattern of the image buffer to the image data stored in the image buffer.
-
公开(公告)号：EP3671575A3
公开(公告)日：2020-09-16
申请号：EP19204904.7
申请日：2019-10-23
Applicant: Samsung Electronics Co., Ltd.
Inventor： LEE, Seoh Yung , KWAK, Youngjun , SON, Changyong , SON, Jinwoo , JUNG, San Gil , CHOI, Chang Kyu , HAN, Jaejoon
Abstract: A neural network processing method and apparatus based on nested bit representation is provided. The processing method includes obtaining first weights for a first layer of a source model of a first layer of a neural network, determining a bit-width for the first layer of the neural network, obtaining second weights for the first layer of the neural network by extracting at least one bit corresponding to the determined bit-width from each of the first weights for the first layer of a source model corresponding to the first layer of the neural network, and processing input data of the first layer of the neural network by executing the first layer of the neural network based on the obtained second weights.
-
公开(公告)号：EP3528181A1
公开(公告)日：2019-08-21
申请号：EP19156720.5
申请日：2019-02-12
Applicant: Samsung Electronics Co., Ltd.
Inventor： SON, Changyong , SON, Jinwoo , JUNG, Sangil , CHOI, Chang Kyu , HAN, Jae-Joon
Abstract: A processing method using a neural network includes generating output maps of a current layer of the neural network by performing a convolution operation between input maps of the current layer and weight kernels of the current layer, determining a lightweight format for the output maps of the current layer based on a distribution of at least a portion of activation data being processed in the neural network, and lightening activation data corresponding to the output maps of the current layer to have a low bit width based on the determined lightweight format.
-
-
-
-
-
-
-
-
-
Search Results
10
records
(took 0.085 seconds)
