Patents by Inventor Soonmin Bae
Soonmin Bae has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11972578Abstract: A method and system for tracking an object in an input video using online training includes a step for training a classifier model by using global pattern matching, and a step for classifying and tracking each target through online training including the classifier model.Type: GrantFiled: August 27, 2021Date of Patent: April 30, 2024Assignee: NAVER CORPORATIONInventors: Myunggu Kang, Dongyoon Wee, Soonmin Bae
-
Patent number: 11610331Abstract: Provided is a method of generating data for estimating a three-dimensional (3D) pose of an object included in an input image, the method including acquiring the input image including at least one moving object, estimating location information about each joint of a plurality of joints of the object included in the input image using a prediction model pretrained to estimate the location information about each joint of the plurality of joints of the object, and generating animation data that represents a movement of the object based on the estimated location information.Type: GrantFiled: December 30, 2020Date of Patent: March 21, 2023Assignee: LINE CORPORATIONInventors: Suntae Kim, Dong Hyun Hwang, Nicolas Monet, Soonmin Bae
-
Publication number: 20210390347Abstract: A method and system for tracking an object in an input video using online training includes a step for training a classifier model by using global pattern matching, and a step for classifying and tracking each target through online training including the classifier model.Type: ApplicationFiled: August 27, 2021Publication date: December 16, 2021Inventors: Myunggu KANG, Dongyoon WEE, Soonmin BAE
-
Publication number: 20210209788Abstract: Provided is a method of generating data for estimating a three-dimensional (3D) pose of an object included in an input image, the method including acquiring the input image including at least one moving object, estimating location information about each joint of a plurality of joints of the object included in the input image using a prediction model pretrained to estimate the location information about each joint of the plurality of joints of the object, and generating animation data that represents a movement of the object based on the estimated location information.Type: ApplicationFiled: December 30, 2020Publication date: July 8, 2021Applicants: NAVER CORPORATION, LINE CorporationInventors: Suntae KIM, Dong Hyun HWANG, Nicolas MONET, Soonmin BAE
-
Patent number: 10348961Abstract: A camera modeling system. The camera modeling system includes: a vanishing point detecting module configured to detect, from at least one reference detection object existing in a detection region, a first vanishing point and a second vanishing point; a view mode determining module configured to determine a camera view mode on the basis of a distance from a first position in the detection region to the first vanishing point and a distance from the first position to the second vanishing point; and a camera modeling module configured to select a camera modeling mode corresponding to the determined camera view mode and perform camera modeling through the selected camera modeling mode.Type: GrantFiled: March 30, 2015Date of Patent: July 9, 2019Assignees: HANWHA PRECISION MACHINERY CO., LTD., HANWHA TECHWIN CO., LTD.Inventors: Soonmin Bae, Hana Hong
-
Patent number: 9978155Abstract: A method and apparatus for counting a number of persons in an image are provided. The person counting apparatus includes: a detector configured to detect a movement area in a top-view distance image, of at least one image frame, providing distance information; an estimator configured to estimate a candidate head area from the movement area based on the distance information; and a determiner configured to determine whether the candidate head area is an actual head of a person based on a similarity to a circle of the candidate head area and a spherical curvature of the candidate head area estimated from the distance information about the candidate head area.Type: GrantFiled: August 24, 2015Date of Patent: May 22, 2018Assignee: Hanwha Techwin Co., Ltd.Inventors: Youngmin Baek, Soonmin Bae
-
Patent number: 9740941Abstract: A method for visualizing loitering objects includes: detecting at least one object determined to have been in a selected area of an input image for a preset time period; obtaining representative still images of each of the detected at least one object in respective time periods during the preset time period; and displaying the representative still images in a time order, or generating a video summary in which images of each of the detected at least one object, respectively included in the representative still images, are displayed together on a single image with indication of the time order.Type: GrantFiled: May 4, 2015Date of Patent: August 22, 2017Assignee: Hanwha Techwin Co., Ltd.Inventors: Soonmin Bae, Mookyung Song
-
Publication number: 20160292890Abstract: A method and apparatus for counting a number of persons in an image are provided. The person counting apparatus includes: a detector configured to detect a movement area in a top-view distance image, of at least one image frame, providing distance information; an estimator configured to estimate a candidate head area from the movement area based on the distance information; and a determiner configured to determine whether the candidate head area is an actual head of a person based on a similarity to a circle of the candidate head area and a spherical curvature of the candidate head area estimated from the distance information about the candidate head area.Type: ApplicationFiled: August 24, 2015Publication date: October 6, 2016Applicant: HANWHA TECHWIN CO., LTD.Inventors: Youngmin BAEK, Soonmin BAE
-
Publication number: 20160117827Abstract: A method for visualizing loitering objects includes: detecting at least one object determined to have been in a selected area of an input image for a preset time period; obtaining representative still images of each of the detected at least one object in respective time periods during the preset time period; and displaying the representative still images in a time order, or generating a video summary in which images of each of the detected at least one object, respectively included in the representative still images, are displayed together on a single image with indication of the time order.Type: ApplicationFiled: May 4, 2015Publication date: April 28, 2016Applicant: HANWHA TECHWIN CO.,LTD.Inventors: Soonmin BAE, Mookyung SONG
-
Publication number: 20150288878Abstract: A camera modeling system. The camera modeling system includes: a vanishing point detecting module configured to detect, from at least one reference detection object existing in a detection region, a first vanishing point and a second vanishing point; a view mode determining module configured to determine a camera view mode on the basis of a distance from a first position in the detection region to the first vanishing point and a distance from the first position to the second vanishing point; and a camera modeling module configured to select a camera modeling mode corresponding to the determined camera view mode and perform camera modeling through the selected camera modeling mode.Type: ApplicationFiled: March 30, 2015Publication date: October 8, 2015Applicant: SAMSUNG TECHWIN CO., LTD.Inventors: Soonmin BAE, Hana HONG
-
Patent number: 8081838Abstract: The present invention provides two-scale tone management of an input image. A method of providing two-scale tone management contains the steps of: separating the input image into a base layer and a detail layer; separating a model image into a base layer and a detail layer; analyzing the input image globally for global contrast; analyzing the input image locally for local contrast; and performing detail preservation of the input image. A system contains a memory and a processor, where the processor is configured by the memory to perform the steps of: separating the input image into a base layer and a detail layer; separating a model image into a base layer and a detail layer; analyzing the input image globally for global contrast; analyzing the input image locally for local contrast; and performing detail preservation of the input image.Type: GrantFiled: March 16, 2007Date of Patent: December 20, 2011Assignee: Massachusetts Institute of TechnologyInventors: Soonmin Bae, Fréderic Durand, Sylvain Paris
-
Patent number: 7945107Abstract: The present invention provides gradient preservation of an input image. A method of providing gradient preservation of an input image, comprises the steps of: computing gradient fields of the input image; computing gradient fields of a remapped image, wherein the remapped image is the input image remapped; comparing a gradient of the remapped image to a predefined range of acceptable values, wherein the acceptable values depend on the input gradient; changing values of the gradient of the remapped image if the values of the gradient of the remapped image are not within the predefined range of acceptable values depending on the input gradient, resulting in x and y components of a modified gradient; and reconstructing the input image using the x and y components of the modified gradient.Type: GrantFiled: March 16, 2007Date of Patent: May 17, 2011Assignee: Massachusetts Institute of TechnologyInventors: Soonmin Bae, Fréderic Durand, Sylvain Paris
-
Publication number: 20080226190Abstract: The present invention provides gradient preservation of an input image. A method of providing gradient preservation of an input image, comprises the steps of: computing gradient fields of the input image; computing gradient fields of a remapped image, wherein the remapped image is the input image remapped; comparing a gradient of the remapped image to a predefined range of acceptable values, wherein the acceptable values depend on the input gradient; changing values of the gradient of the remapped image if the values of the gradient of the remapped image are not within the predefined range of acceptable values depending on the input gradient, resulting in x and y components of a modified gradient; and reconstructing the input image using the x and y components of the modified gradient.Type: ApplicationFiled: March 16, 2007Publication date: September 18, 2008Applicant: Massachusetts Institute of TechnologyInventors: Soonmin Bae, Frederic Durand, Sylvain Paris
-
Publication number: 20080226168Abstract: The present invention provides two-scale tone management of an input image. A method of providing two-scale tone management contains the steps of: separating the input image into a base layer and a detail layer; separating a model image into a base layer and a detail layer; analyzing the input image globally for global contrast; analyzing the input image locally for local contrast; and performing detail preservation of the input image. A system contains a memory and a processor, where the processor is configured by the memory to perform the steps of: separating the input image into a base layer and a detail layer; separating a model image into a base layer and a detail layer; analyzing the input image globally for global contrast; analyzing the input image locally for local contrast; and performing detail preservation of the input image.Type: ApplicationFiled: March 16, 2007Publication date: September 18, 2008Applicant: Massachusetts Institute of TechnologyInventors: Soonmin Bae, Frederic Durand, Sylvain Paris