-
公开(公告)号：US20230177083A1
公开(公告)日：2023-06-08
申请号：US18072268
申请日：2022-11-30
Inventor： Joon Soo KIM , Joung Il YUN , Geun Taek LIM , Won JO , Yu Kyung CHOI , Yu Jin HWANG
IPC: G06F16/783 , G06T7/194 , G06V10/74 , G06V20/40
CPC classification number: G06F16/783 , G06T7/194 , G06V10/761 , G06V20/47
Abstract: Disclosed herein method and apparatus for simultaneous video retrieval and alignment. According to an embodiment of the present disclosure, there is provided a method for retrieving a video. The method comprising: detecting a section of interest in a query video that is a retrieval request video; producing one or more frame-level descriptor and a video-level descriptors for the query video by using key frames within the detected section of interest; and retrieving a reference video corresponding to the query video based on the frame-level descriptor and the video-level descriptor for the query video and one or more frame-level descriptor and a video-level descriptor for each of reference videos stored in a database.
-
Patent Agency Ranking
公开(公告)号：US20190332882A1
公开(公告)日：2019-10-31
申请号：US16383840
申请日：2019-04-15
Inventor： Joon Soo KIM , Gwang Soon LEE
Abstract: Disclosed is an apparatus and method of correcting 3D image distortion. A method of correcting 3D image distortion according to the present disclosure includes: receiving an input image that contains a predetermined first pattern; extracting a characteristic value related to the first pattern from the input image; and updating the input image on the basis of the extracted characteristic value.
-
3.发明公开公开(公告)号：US20240020335A1
公开(公告)日：2024-01-18
申请号：US18298208
申请日：2023-04-10
Inventor： Joon Soo KIM , Joung Il YUN , Geun Taek LIM , Won JO , Yu Kyung CHOI
IPC: G06F16/732 , G06F16/583 , G06F16/783 , G06V10/771 , G06V10/82
CPC classification number: G06F16/7328 , G06F16/583 , G06F16/783 , G06V10/771 , G06V10/82
Abstract: The present disclosure discloses an image/video feature descriptor extraction method and apparatus for image/video search. An image/video feature descriptor extraction method may comprise obtaining a search requested image/video from a user, performing spatio-temporal transformation on the search requested image/video, extracting a feature map from the image/video subjected to the spatio-temporal transformation, and extracting an image/video feature descriptor by applying TNIP (Temporal Nested Invariance Pooling) to the feature map.
-
公开(公告)号：US20210349532A1
公开(公告)日：2021-11-11
申请号：US17307730
申请日：2021-05-04
Inventor： Hyun Cheol KIM , Joon Soo KIM , Gwang Soon LEE
Abstract: Disclosed herein a device tracking gaze and method therefor. The device includes: an image acquisition unit configured to obtain an eyeball image; a pupil detection unit configured to detect a center of pupil by using the eyeball image; a virtual corneal reflection light position generator configured to process the eyeball image so that a virtual corneal reflection light is located at a predetermined point in the eyeball image; and a PCVR vector generator configured to generate a pupil center virtual reflection vector (PCVR vector) based on a position of the pupil center and a position of the virtual corneal reflection light.
-
5.发明公开公开(公告)号：US20240282049A1
公开(公告)日：2024-08-22
申请号：US18443936
申请日：2024-02-16
Inventor： Joon Soo KIM
CPC classification number: G06T15/506 , G06T7/11 , G06T7/73 , G06T2207/20081 , G06T2207/20084
Abstract: A device, method, and system for restoring a 3D shape based on multiple light sources are disclosed. A method performed by a device may include obtaining a set of measured images of a specimen by photographing the specimen under different lighting conditions; defining a first three-dimensional grid within a 3D reconstruction area including all or part of the specimen; obtaining a first set of coordinate values for calculating a scattering potential value within a first 3D grid or a frequency grid region corresponding to the first 3D grid; and obtaining a 3D refractive index distribution of the specimen based on the output data obtained through an artificial neural network model, and the artificial neural network model is trained based on differences between the set of predicted images and the set of the measured images.
-
公开(公告)号：US20210360218A1
公开(公告)日：2021-11-18
申请号：US17321213
申请日：2021-05-14
Applicant: Electronics and Telecommunications Research Institute , SOGANG UNIVERSITY RESEARCH & BUSINESS DEVELOPMENT FOUNDATION
Inventor： Joon Soo KIM , Kug Jin YUN , Jun Young JEONG , Suk Ju KANG , Jung Hee KIM , Yeo Hun YUN
IPC: H04N13/128 , H04N13/167 , H04N13/282
Abstract: Disclosed herein is a method for rectifying a 2D multi-view image. The method for rectifying a 2D multi-view image according to an embodiment of the present disclosure may include: detecting uniformly the at least one feature point in each region unit distinguished by considering a distribution of feature points of a plurality of input images: removing an error of the least one feature point; determining a corresponding pair for a vertical or horizontal direction of the at least one feature point; by considering an arrangement relationship of the plurality of input images, projecting the at least one feature point onto a projection plane; determining a disparity error for a corresponding pair for the at least one feature point that is projected onto the projection plane; and, by considering the disparity error, performing image rectification based on the at least one feature point.
-
-
-
-
-
Search Results
6
records
(took 0.021 seconds)
