公开(公告)号：US20230222781A1

公开(公告)日：2023-07-13

申请号：US17970907

申请日：2022-10-21

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Insoo KIM ,  Kikyung KIM ,  Seungju HAN ,  Jiwon BAEK ,  Jaejoon HAN

IPC: G06V10/80 ,  G06V10/77 ,  G06V10/82

CPC classification number: G06V10/806 ,  G06V10/7715 ,  G06V10/82

Abstract: A method and apparatus for object recognition are provided. A processor-implemented method includes extracting feature maps including local feature representations from an input image, generating a global feature representation corresponding to the input image by fusing the local feature representations, and performing a recognition task on the input image based on the local feature representations and the global feature representation.

公开(公告)号：US20210166071A1

公开(公告)日：2021-06-03

申请号：US16913205

申请日：2020-06-26

Applicant: Samsung Electronics Co., Ltd.

Inventor： Seong-Jin PARK ,  Insoo KIM ,  Seungju HAN ,  Jiwon BAEK ,  Ju Hwan SONG ,  Jaejoon HAN

IPC: G06K9/62 ,  G06N3/04 ,  G06N3/08

Abstract: A processor-implemented neural network method includes: determining, using a neural network, a feature vector based on a training image of a first class among a plurality of classes; determining, using the neural network, plural feature angles between the feature vector and class vectors of other classes among the plurality of classes; determining a margin based on a class angle between a first class vector of the first class and a second class vector of a second class, among the class vectors, and a feature angle between the feature vector and the first class vector; determining a loss value using a loss function including an angle with the margin applied to the feature angle and the plural feature angles; and training the neural network by updating, based on the loss value, either one or both of one or more parameters of the neural network and one or more of the class vectors.

公开(公告)号：US20200066280A1

公开(公告)日：2020-02-27

申请号：US16519757

申请日：2019-07-23

Applicant: Samsung Electronics Co., Ltd.

Inventor： Kyuhong KIM ,  Insoo KIM ,  Dohwan LEE ,  Hana LEE

IPC: G10L17/04 ,  G10L17/00 ,  G06N3/04 ,  G10L15/16

Abstract: A speaker authentication method and apparatus may extract input speaker features corresponding to a plurality of frames of an input speech of an object, estimate discriminable speaker sections corresponding to the plurality of frames, and dynamically match the input speaker features to pre-enrolled enrolled speaker features based on the discriminable speaker section.

公开(公告)号：US20250157205A1

公开(公告)日：2025-05-15

申请号：US18934024

申请日：2024-10-31

Applicant: Samsung Electronics Co., Ltd. ,  POSTECH RESEARCH AND BUSINESS DEVELOPMENT FOUNDATION

Inventor： Jae seok CHOI ,  Manjin KIM ,  SangHyun KIM ,  Woohyeok KIM ,  Insoo KIM ,  Min Jung LEE ,  Jaesung RIM ,  Deun Sol JUNG ,  Minsu CHO ,  Sunghyun CHO

IPC: G06V10/80 ,  G06T7/20 ,  G06V10/44 ,  G06V10/62 ,  G06V10/74 ,  G06V10/771

Abstract: A method and apparatus with image enhancement using a base image are provided. The method includes: extracting, by a spatial feature extractor, intra-image spatial feature representations from respective image frames of a burst image set; generating inter-image temporal feature representations based on a local similarity between the spatial feature representations; determining temporal-spatial feature representations of the respective image frames by fusing the spatial feature representations with the temporal feature representations; selecting a base image frame from among the image frames based on the temporal-spatial feature representations; and generating an enhanced image by performing an image enhancement operation on the burst image set based on the base image frame.

公开(公告)号：US20230366982A1

公开(公告)日：2023-11-16

申请号：US18354179

申请日：2023-07-18

Applicant: Samsung Electronics Co., Ltd.

Inventor： Insoo KIM ,  Seungju HAN ,  Jaejoon HAN

IPC: G01S7/41 ,  G06F21/44

CPC classification number: G01S7/412 ,  G06F21/44

Abstract: A processor-implemented radar data recognition method including generating a plurality of pieces of input data, with respectively different dimension configurations, based on radar data of an object; and outputting a recognition result of the object based on the generated plurality of pieces of input data using a recognition model.

公开(公告)号：US20220058377A1

公开(公告)日：2022-02-24

申请号：US17208048

申请日：2021-03-22

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jiwon BAEK ,  Seong-Jin PARK ,  Seungju HAN ,  Insoo KIM ,  Jaejoon HAN

IPC: G06K9/00 ,  G06T9/00 ,  G06K9/62 ,  G06T7/73 ,  G06T3/60 ,  G06T3/40 ,  G06N3/04 ,  G06N3/08

Abstract: A processor-implemented facial image generating method includes: determining a first feature vector associated with a pose and a second feature vector associated with an identity by encoding an input image including a face; determining a flipped first feature vector by flipping the first feature vector with respect to an axis in a corresponding space; determining an assistant feature vector based on the flipped first feature vector and rotation information corresponding to the input image; determining a final feature vector based on the first feature vector and the assistant feature vector; and generating an output image including a rotated face by decoding the final feature vector and the second feature vector based on the rotation information.