-
公开(公告)号：KR101678453B1
公开(公告)日：2016-11-23
申请号：KR1020140098045
申请日：2014-07-31
Applicant: 한국과학기술연구원
Abstract: 본명세서의일 실시예에따른이미지처리방법은전체이미지를나타내는학습이미지를서브이미지들로분할하고, 서브이미지들의관계를연상메모리모델에서학습하는학습단계; 및전체이미지의부분을나타내는질의이미지가입력된경우, 학습된연상메모리모델을이용하여질의이미지를전체이미지로재구성하는복원단계를포함할수 있다.
Abstract translation: 根据本发明实施例的图像处理方法包括：学习步骤，将表示整个图像的学习图像分成子图像，并且在关联存储器模型中学习子图像的关系; 并且当输入代表整个图像的一部分的查询图像时，使用所学习的关联存储器模型将查询图像重建为整个图像。
-
公开(公告)号：KR101671391B1
公开(公告)日：2016-11-02
申请号：KR1020150096624
申请日：2015-07-07
Applicant: 한국과학기술연구원
IPC: G06T5/00
CPC classification number: G06T5/003 , G06T2207/10016 , G06T2207/20201
Abstract: 레이어블러모델에기반한비디오디블러링방법은, 블러비디오프레임을수신하면, 카메라의노출시간동안의레이어의결합으로이루어진이미지들을이용하여, 각프레임의레이어별잠상(latent image), 물체의움직임및 마스크를추정하는단계; 추정된각 프레임의레이어별잠상, 물체의움직임및 마스크를레이어블러모델에적용하여블러프레임을생성하는단계; 생성된블러프레임과수신된블러비디오프레임을비교하는단계; 및상기생성된블러프레임과상기수신된블러비디오프레임이일치하는경우, 상기추정된각 프레임의레이어별물체의움직임및 마스크를기초로, 최종잠상을출력하는단계를포함한다. 이에따라, 블러된영상을노출도중의전경과배경의결합으로이루어진이미지들의중첩으로모델링함으로써, 경계에서의가리어짐이표현가능하여물체의경계에서보다정확한디블러링결과를얻어낼수 있다.
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，对象运动和掩模， 将每个帧中的每个层的估计潜像，对象运动和掩模施加到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，并且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。
-
公开(公告)号：KR1020160125088A
公开(公告)日：2016-10-31
申请号：KR1020150055815
申请日：2015-04-21
Applicant: 한국과학기술연구원
Inventor： 이석 , 김재헌 , 김익재 , 이택진 , 김현덕 , 정현수 , 최지호 , 김형석 , 경규형 , 정원교 , 신정훈 , 엄재홍 , 박진성 , 유현복 , 팽혜선 , 이병호
Abstract: 조립유닛및 이를포함하는조립완구가개시된다. 상기조립완구는양전극과음전극이결합스터드에교번하여배열되는제1전원라인과상기결합스터드간의무게중심점에마련되는제1데이터라인을포함하는베이스블록및 상기베이스블록에결합하여상기제1전원라인으로부터전원을공급받고, 상기제1데이터라인을통하여데이터통신을수행하는서브유닛을포함한다.
Abstract translation: 公开了一种组装单元和包括该组装单元的组装玩具。 组装玩具是正电极和负电极被耦合到耦合到所述基块和所述第一电源线的基块包括交替地一个螺栓和所述第一电源线和布置在重力的联接柱之间的中心设置在第一数据线 以及用于通过第一数据线执行数据通信的子单元。
-
-
公开(公告)号：KR1020160061081A
公开(公告)日：2016-05-31
申请号：KR1020140163534
申请日：2014-11-21
Applicant: 한국과학기술연구원
IPC: G06T7/00
CPC classification number: G06K9/00228 , G06K9/00208 , G06K9/00268 , G06K9/00302 , G06T7/00 , G06T7/0002 , G06T7/0004 , G06T7/337 , G06T7/344 , G06T7/37 , G06T7/536
Abstract: 얼굴표정정규화를통한얼굴인식방법은, 3차원얼굴데이터베이스를이용하여, 입력되는 2차원얼굴영상을 3차원얼굴모델로정합하는단계; 상기 3차원얼굴데이터베이스로부터학습된무표정파라미터를이용하여, 상기 3차원얼굴모델을무표정의 3차원얼굴모델로정규화하는단계; 상기무표정의 3차원얼굴모델을무표정의 2차원얼굴영상으로변환하는단계; 및 2차원얼굴데이터베이스로부터상기무표정의 2차원얼굴영상을인식하는단계를포함한다. 이에따라, 정보의손실없이신뢰성높은얼굴인식을수행할수 있다.
Abstract translation: 使用面部表情归一化的面部识别方法包括以下步骤：通过使用三维面部数据库将输入的二维面部图像拟合成三维面部模型; 通过使用从三维面部数据库中学习的中性表达参数将三维面部模型归一化为中性表达三维面部模型; 将中性表情三维面部模型转化为中性表情二维面部图像; 并从三维面部数据库识别中性表达二维面部图像。 因此，可以可靠地执行面部识别处理而不丢失信息。
-
Patent Agency Ranking
公开(公告)号：KR1020140060437A
公开(公告)日：2014-05-20
申请号：KR1020120126944
申请日：2012-11-09
Applicant: 한국과학기술연구원
IPC: G01S3/78
CPC classification number: G06F3/0325 , G06F3/011 , G06K9/2036
Abstract: According to the present invention, an apparatus for wirelessly recognizing a location of a moving object in real time includes a lighting portion including a lighting unit for emitting a plurality of optical signals spatially separated in different patterns; an electronic tattoo portion which is attached on the moving object, and includes at least one electronic tattoo unit including a solar cell for detecting each optical signal emitted from the lighting portion, a wireless antenna for wirelessly transmitting the optical signals detected by the solar cell, and a controller for controlling the transmission of the optical signals; and a location recognition portion for recognizing the location of the moving object from the optical signals wirelessly transmitted from the electronic tattoo portion.
Abstract translation: 根据本发明，实时无线地识别移动物体的位置的装置包括：照明部，其包括用于发射以不同图案空间分离的多个光信号的照明单元; 电子纹身部分，其附着在移动物体上，并且包括至少一个电子纹身单元，其包括用于检测从照明部分发射的每个光信号的太阳能电池，用于无线发送由太阳能电池检测的光信号的无线天线， 以及控制器，用于控制光信号的传输; 以及位置识别部分，用于从从电子纹身部分无线发送的光信号中识别移动物体的位置。
-
公开(公告)号：KR1020140059076A
公开(公告)日：2014-05-15
申请号：KR1020120125704
申请日：2012-11-07
Applicant: 한국과학기술연구원
IPC: G06Q50/10
CPC classification number: G06F3/0482 , G06F3/0481 , G06F3/04817 , G06F17/30274 , G06K9/00308 , G06N3/006 , G06N3/126 , H04N1/00442
Abstract: The present invention relates to an apparatus and a method for generating a cognitive avatar, which repetitively perform a process of enabling a user to select imaged which are recognized to be similar, among face images with various features stored in a database where multiple face images are classified into multiple feature groups and are stored. The apparatus and the method of the present invention can easily express a natural avatar with a similar feature to a target face without an additional analysis or reinterpretation process for the target face by accessing cognitively and scientifically based on the selection of the user which is performed repetitively from the above process, and by generating the avatar corresponding to the target face which the user wants to generate as an avatar.
Abstract translation: 本发明涉及一种用于产生认知头像的装置和方法，其重复地执行允许用户在被存储在数据库中的各种特征的面部图像中选择被识别为相似的成像的处理，其中多个脸部图像是 分为多个特征组并存储。 本发明的装置和方法可以基于对重复执行的用户的选择，通过认知和科学地访问具有与目标面部相似特征的自然头像，而无需对目标面部进行额外的分析或重新解释处理 并且通过生成与用户想要生成的目标面对应的化身作为头像。
-
公开(公告)号：KR1020140037741A
公开(公告)日：2014-03-27
申请号：KR1020130025261
申请日：2013-03-08
Applicant: 한국과학기술연구원
CPC classification number: G06F3/0346 , G06F3/0308 , G06F3/0325 , G06F2203/0384
Abstract: A glove apparatus for hand gesture recognition and interaction according to the present invention is characterized by including a lighting unit including k number (k is a natural number equal to or greater than 1) of lighting units for irradiating n number of optical signals (n is a natural number equal to or greater than 1) spatially-split in mutually different patterns, a glove marker unit having the shape of a glove and having m number of markers (m is a natural number equal to or greater than 1) that are attached to a plurality of areas of the glove to detect the optical signals, respectively, irradiated from the lighting unit, and a location information outputting unit for recognizing information of locations of the markers based on the optical signals received from the glove markers and for wirelessly transmitting the information of the location of each recognized marker to the display device. [Reference numerals] (100) Lighting module; (200) Glove marker module; (300) Location information outputting module; (400) Haptic function performing module
Abstract translation: 根据本发明的用于手势识别和交互的手套装置的特征在于包括：用于照射n个光信号的照明单元的k个数（k为自然数等于或大于1）的照明单元（n为 等于或大于1的自然数）在空间上分开相互不同的图案，具有手套形状并且具有m个标记的手套标记单元（m是等于或大于1的自然数），附着 到手套的多个区域以分别从照明单元照射的光信号和位置信息输出单元，用于基于从手套标记接收的光信号识别标记的位置的信息，并且用于无线发送 将每个识别的标记的位置的信息显示给显示装置。 （附图标记）（100）照明模块; （200）手套标记模块; （300）位置信息输出模块; （400）触觉功能执行模块
-
公开(公告)号：KR101360949B1
公开(公告)日：2014-02-12
申请号：KR1020120031214
申请日：2012-03-27
Applicant: 한국과학기술연구원
CPC classification number: G01C21/365
Abstract: 본 발명의 일 실시예로서, 차량의 전방에 목적지까지의 주행정보를 조사(project)하기 위한 방법 및 장치가 제공된다. 차량의 전방에 목적지까지의 주행정보를 조사하기 위한 방법은 주행정보를 수집하고, 수집된 주행정보에 기초하여, 주행정보 조사 장치를 제어하는 단계를 포함할 수 있다. 주행정보 조사 장치는 주행정보를 표시하는 주행정보 표시부와 상기 주행정보 표시부에 대하여 광원으로서 이용되는 조명부를 포함할 수 있다.
본 발명에 따르면, 차량 운행자는 종래의 네비게이션 장치들의 액정화면으로 시선을 이동시킬 필요 없이, 운행중인 차량 전방에 표시되는 주행 정보를 통하여 주행 정보 등을 쉽사리 인지할 수 있고, 이를 통해 더욱 안전한 운행이 보장될 수 있다.-
公开(公告)号：KR1020130131122A
公开(公告)日：2013-12-03
申请号：KR1020120054944
申请日：2012-05-23
Applicant: 한국과학기술연구원
CPC classification number: G06F3/0346 , G06F3/0481 , G06F3/0488 , G06F3/04812 , G06F1/1694 , G06F3/04815 , G06F2203/0381
Abstract: The present invention relates to a method for controlling a three dimensional virtual cursor using a portable electronic device comprising the step of detecting one or more of the movement and a touch input of the portable electronic device through a sensor mounted at the portable electronic device; and the step of converting at least one of the detected movement and the touch input of the portable electronic device into a cursor controlling signal controlling the operation of a cursor at a three dimensional space for outputting. Thus the present invention enables a user to conveniently control a three dimensional virtual cursor without the limit of the space or time whenever needed by using the portable device carried by the user. [Reference numerals] (AA) START;(BB) END;(S110) Sensing at least one of the movement and the touch input of a portable electronic device through a sensor mounted at the portable electronic device;(S120) Converting at least one of the sensed movement and the touch input of the portable electronic device into a cursor control signal at a 3D virtual space before outputting
Abstract translation: 本发明涉及一种使用便携式电子设备控制三维虚拟光标的方法，包括通过安装在便携式电子设备上的传感器检测便携式电子设备的移动和触摸输入中的一个或多个的步骤; 以及将便携式电子设备的检测到的移动和触摸输入中的至少一个转换为控制三维空间中的光标的操作以输出的光标控制信号的步骤。 因此，本发明使用户能够方便地控制三维虚拟光标，而无需使用用户携带的便携式设备的需要时的空间或时间的限制。 （AA）START;（BB）END;（S110）通过安装在便携式电子设备上的传感器感测便携式电子设备的移动和触摸输入中的至少一个;（S120）将至少一个 的感测移动和便携式电子设备的触摸输入到输出之前的3D虚拟空间处的光标控制信号
-
-
-
-
-
-
-
-
-
Search Results
149
records
(took 0.031 seconds)
