公开(公告)号：US20210166486A1

公开(公告)日：2021-06-03

申请号：US17080296

申请日：2020-10-26

Applicant: Samsung Electronics Co., Ltd.

Inventor： Yongsung KIM ,  Daehyun BAN ,  Dongwan LEE ,  Hongpyo LEE ,  Lei ZHANG

IPC: G06T19/00 ,  G06K9/00 ,  G06T7/50 ,  G06K9/62 ,  G06T7/70 ,  G06N3/08

Abstract: An electronic apparatus is provided. The electronic apparatus includes a display, a camera configured to capture a rear of the electronic apparatus facing a front of the electronic apparatus in which the display displays an image, and a processor configured to render a virtual object based on the image captured by the camera, based on a user body being detected from the captured image, estimate a plurality of joint coordinates with respect to the detected user body using a pre-trained learning model, generate an augmented reality image using the estimated plurality of joint coordinates, the rendered virtual object, and the captured image, and control the display to display the generated augmented reality image, wherein the processor is configured to identify whether the user body touches the virtual object based on the plurality of estimated joint coordinates, and change a transmittance of the virtual object based on the touch being identified.

公开(公告)号：US20230206617A1

公开(公告)日：2023-06-29

申请号：US18068209

申请日：2022-12-19

Applicant: Samsung Electronics Co., Ltd.

Inventor： Daehyun BAN ,  Daehun KIM ,  Yongsung KIM ,  Dongwan LEE ,  Juyoung LEE

IPC: G06V10/82 ,  G06T5/20 ,  G06T7/50 ,  G06T7/70 ,  G06V10/764 ,  H04N19/59

CPC classification number: G06V10/82 ,  G06T5/20 ,  G06T7/50 ,  G06T7/70 ,  G06V10/764 ,  H04N19/59

Abstract: An apparatus is provided. The apparatus includes an input/output interface configured to receive an image and output a result, a memory storing one or more instructions for processing the image by using a convolutional neural network, and a processor configured to process the image by executing the one or more instructions, wherein the convolutional neural network (CNN) may include one or more spatial transformation modules, and the spatial transformation module may include a spatial transformer configured to apply a spatial transform to first input data that is the image or an output of a previous spatial transformation module, by using a spatial transformation function, a first convolutional layer configured to perform a convolution operation between the first input data to which the spatial transform is applied and a first filter, and a spatial inverse transformer configured to apply a spatial inverse transform to an output of the first convolutional layer.

公开(公告)号：US20230169679A1

公开(公告)日：2023-06-01

申请号：US17963419

申请日：2022-10-11

Applicant: Samsung Electronics Co., Ltd.

Inventor： Yongsung KIM ,  Hyunsoo CHOI ,  Daehyun BAN ,  Dongwan LEE ,  Juyoung LEE

IPC: G06T7/73 ,  G06T3/60

CPC classification number: G06T7/73 ,  G06T3/60 ,  G06T2207/30196 ,  G06T2207/20081

Abstract: A server for pose estimation of a person and an operating method of the server are provided. The operating method includes obtaining an original image including a person, generating a plurality of input images by rotating the original image, obtaining first pose estimation results respectively corresponding to the plurality of input images, by inputting the plurality of input images to a pose estimation model, applying weights to the first pose estimation results respectively corresponding to the plurality of input images, and obtaining a second pose estimation result, based on the first pose estimation results to which the weights are applied, wherein the first pose estimation results and the second pose estimation result each include data indicating main body parts of the person.

公开(公告)号：US20220375032A1

公开(公告)日：2022-11-24

申请号：US17745192

申请日：2022-05-16

Applicant: Samsung Electronics Co., Ltd.

Inventor： Daehyun BAN ,  Yongsung KIM ,  Dongwan LEE ,  Juyoung LEE

IPC: G06T3/40 ,  G06V10/77 ,  G06T5/30 ,  G06T5/50

Abstract: A method of operating an image processing apparatus is provided. The method includes generating a first feature map by performing a convolution operation between a first image and a first kernel group, generating a second feature map by performing a convolution operation between the first image and a second kernel group, generating a first combination map based on the first feature map, generating a second combination map based on the first feature map and the second feature map, generating a second image based on the first combination map and the second combination map, and generating a reconstructed image of the first image, based on the second image and the first image, and generating a high-resolution image of the first image by inputting the reconstructed image to an upscaling model.