-
1.发明公开公开(公告)号：US20240185041A1
公开(公告)日：2024-06-06
申请号：US18450833
申请日：2023-08-16
Applicant: KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY , SOGANG UNIVERSITY RESEARCH & BUSINESS DEVELOPMENT FOUNDATION
Inventor： Junghyun CHO , Igjae KIM , Unsang PARK , Haetsal LEE
IPC: G06N3/0464 , G06N3/048
CPC classification number: G06N3/0464 , G06N3/048
Abstract: The present disclosure relates to a technology for skeleton-based action recognition based on a graph convolutional network, in which an action processing device receives a frame including a skeleton with respect to actions of an object, extracts spatiotemporal features with respect to the skeleton by using a rank adjacency matrix in which a distance between one node and another node and an adjacency ranking are considered, merges an object and vertices in the input frame based on the extracted spatiotemporal features, and performs a classification task.
-
公开(公告)号：US20220300591A1
公开(公告)日：2022-09-22
申请号：US17535575
申请日：2021-11-25
Applicant: KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY
Inventor： Ig Jae KIM , Gi Pyo NAM , Haesol PARK , Junghyun CHO , Je Hyeong HONG
Abstract: Embodiments relate to a user authentication device configured to detect a face region in a target object image including at least part of a face of a target object, recognize masked or unmasked in the face region, extract target object characteristics data from the face region of the target object image, call reference data and authenticate if the target object is a registered device user based on the called reference data and the target object characteristics data. The reference data is generated from an unmasked image of the registered device user.
-
3.发明申请公开(公告)号：US20180268207A1
公开(公告)日：2018-09-20
申请号：US15985173
申请日：2018-05-21
Applicant: Korea Institute of Science and Technology
Inventor： Ig Jae KIM , Heeseung CHOI , Sungyeon PARK , Junghyun CHO
Abstract: A method for automatic facial impression transformation includes extracting landmark points for elements of a target face whose facial impression is to be transformed as well as distance vectors respectively representing distances of the landmark points, comparing the distance vectors to select a learning data set similar to the target face from a database, extracting landmark points and distance vectors from the learning data set, transforming a local feature of the target face based on the landmark points of the learning data set and score data for a facial impression, and transforming a global feature of the target face based on the distance vectors of the learning data set and the score data for the facial impression. Accordingly, a facial impression may be transformed in various ways while keeping an identity of a corresponding person.
-
公开(公告)号：US20170116705A1
公开(公告)日：2017-04-27
申请号：US15069095
申请日：2016-03-14
Applicant: Korea Institute of Science and Technology
Inventor： Ig Jae KIM , Heeseung CHOI , Sungyeon PARK , Junghyun CHO
CPC classification number: G06T3/0093 , G06K9/00281 , G06K9/00302 , G06T11/00 , G06T15/00 , H04N7/147 , H04N7/157
Abstract: A method for automatic facial impression transformation includes extracting landmark points for elements of a target face whose facial impression is to be transformed as well as distance vectors respectively representing distances of the landmark points, comparing the distance vectors to select a learning data set similar to the target face from a database, extracting landmark points and distance vectors from the learning data set, transforming a local feature of the target face based on the landmark points of the learning data set and score data for a facial impression, and transforming a global feature of the target face based on the distance vectors of the learning data set and the score data for the facial impression. Accordingly, a facial impression may be transformed in various ways while keeping an identity of a corresponding person.
-
5.发明公开公开(公告)号：US20230177880A1
公开(公告)日：2023-06-08
申请号：US17973140
申请日：2022-10-25
Applicant: KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY
Inventor： Junghyun CHO , Haksub KIM , Gi Pyo NAM , Hyungjoo JUNG , Ig Jae KIM , Soyoung KIM
CPC classification number: G06V40/20 , G06V10/82 , G06T7/20 , G06T2207/30241
Abstract: Disclosed is a device and method for inferring a correlation between objects through image recognition. The device for inferring a correlation between objects through image recognition according to an embodiment comprises a communicator and an interaction inferencer configured to select a main object interacting with a target object at a predetermined distance in the input image or generate a social graph including the main object and the target object.
-
Patent Agency Ranking
公开(公告)号：US20230169568A1
公开(公告)日：2023-06-01
申请号：US17673965
申请日：2022-02-17
Applicant: KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY
Inventor： Juyoun PARK , Ig Jae KIM , Junghyun CHO , Heeseung CHOI , Haksub KIM , Gi Pyo NAM , Haesol PARK , Hyungjoo JUNG
IPC: G06Q30/06
CPC classification number: G06Q30/0631 , G06Q30/0633
Abstract: A kiosk for providing a recommendation service according to an embodiment displays an orderer's past ordered product as a recommended product on the screen of the kiosk, the past ordered product read based on a similarity calculation result between a current input attribute representing a contextual feature of a current order status and a past input attribute stored in memory.
-
7.发明申请公开(公告)号：US20220206668A1
公开(公告)日：2022-06-30
申请号：US17564226
申请日：2021-12-29
Applicant: KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY
Inventor： Junghyun CHO , Ig Jae KIM , Junyong CHOI , Seok Yeong LEE
IPC: G06F3/04815 , G06T15/20
Abstract: Embodiments relate to a dynamic image capturing method and apparatus using an arbitrary viewpoint image generation technology, in which an image of background content displayed on a background content display unit or an image of background content implemented in a virtual space through a chroma key screen, having a view matching to a view of seeing a subject at a viewpoint of a camera is generated, and a final image including the image of the background content and a subject area is obtained.
-
公开(公告)号：US20220036066A1
公开(公告)日：2022-02-03
申请号：US17104404
申请日：2020-11-25
Applicant: KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY
Inventor： Junghyun CHO , Hyunwoo CHO , Haesol PARK , Ig Jae KIM
Abstract: Disclosed are an X-RAY image reading support method including the steps of acquiring a target X-RAY image photographed by transmitting or reflecting X-RAY in a reading space in which an object to be read is disposed; applying the target X-RAY image to a reading model that extracts features from an input image; and identifying the object to be read as an object corresponding to a classified class when the object to be read is classified as a set class based on a first feature set extracted from the target X-RAY image, and an X-RAY image reading support system performing the method.
-
9.发明申请公开(公告)号：US20180061018A1
公开(公告)日：2018-03-01
申请号：US15630688
申请日：2017-06-22
Applicant: KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY
Inventor： Byeongjoo AHN , Ig Jae KIM , Junghyun CHO
Abstract: A method of multi-view deblurring for 3-dimensional (3D) shape reconstruction includes: receiving images captured by multiple synchronized cameras at multiple viewpoints; performing iteratively estimation of depth map, latent image, and 3D motion at each viewpoint for the received images; determining whether image deblurring at each viewpoint is completed; and performing 3D reconstruction based on final depth maps and latent images at each viewpoint. Accordingly, it is possible to achieve accurate deblurring and 3D reconstruction even from any motion blurred images.
-
10.发明申请METHOD FOR DEBLURRING VIDEO USING MODELING BLURRED VIDEO WITH LAYERS, RECORDING MEDIUM AND DEVICE FOR PERFORMING THE METHOD 有权使用模拟视频的方式来切换视频的方法，记录介质和用于执行方法的设备公开(公告)号：US20170011494A1
公开(公告)日：2017-01-12
申请号：US15191040
申请日：2016-06-23
Applicant: Korea Institute of Science and Technology
Inventor： Byeongjoo AHN , Ig Jae KIM , Junghyun CHO
CPC classification number: G06T5/003 , G06T2207/10016 , G06T2207/20201
Abstract: A video deblurring method based on a layered blur model includes estimating a latent image, an object motion and a mask for each layer in each frame using images consisting of a combination of layers during an exposure time of a camera when receiving a blurred video frame, applying the estimated latent image, object motion and mask for each layer in each frame to the layered blur model to generate a blurry frame, comparing the generated blurry frame and the received blurred video frame, and outputting a final latent image based on the estimated object motion and mask for each layer in each frame, when the generated blurry frame and the received blurred video frame match. Accordingly, by modeling a blurred image as an overlap of images consisting of a combination of foreground and background during exposure, more accurate deblurring results at object boundaries can be obtained.
Abstract translation: 基于分层模糊模型的视频去模糊方法包括：在接收到模糊视频帧时，在相机的曝光时间期间，使用由层的组合构成的图像来估计每个帧中的每个层的潜像，物体运动和掩模， 将每个帧中的每个层的估计潜像，对象运动和掩模应用于分层模糊模型以生成模糊帧，比较生成的模糊帧和接收到的模糊视频帧，并基于估计对象输出最终潜像 当生成的模糊帧和接收的模糊视频帧匹配时，每个帧中的每个层的运动和掩码。 因此，通过将模糊图像建模为在曝光期间由前景和背景的组合组成的图像的重叠，可以获得对象边界处的更准确的去模糊结果。
-
-
-
-
-
-
-
-
-
Search Results
15
records
(took 0.025 seconds)
