-
公开(公告)号：KR101735130B1
公开(公告)日：2017-05-15
申请号：KR1020120146013
申请日：2012-12-14
Applicant: 한국전자통신연구원
Abstract: 본발명은 3차원얼굴복원장치를개시한다. 3차원얼굴복원장치는입력영상의깊이정보와얼굴을나타내는 3차원템플릿을이용하여입력영상에서사용자의얼굴영역을추출하는영역추출부; 및상기사용자의얼굴영역과관련된시간에따른누적깊이값을고려하여 3차원얼굴형상을복원하는형상복원부를포함한다.
Abstract translation: 本发明公开了一种三维人脸重建装置。 该三维人脸重建设备包括：区域提取单元，用于使用输入图像和表示脸部的三维模板的深度信息从输入图像中提取用户的脸部区域; 以及形状恢复单元，用于通过考虑根据与用户的面部区域相关联的时间的累积深度值来恢复三维面部形状。
-
公开(公告)号：KR1020170029204A
公开(公告)日：2017-03-15
申请号：KR1020150126228
申请日：2015-09-07
Applicant: 한국전자통신연구원
CPC classification number: B29C64/386 , B29C64/00 , B29C64/40 , B33Y10/00 , B33Y30/00 , B33Y50/02 , G05B2219/49019
Abstract: 3D 프린터의출력공간보다큰 객체를 3D 프린팅하기위한방법및 이를이용한장치가개시된다. 본발명에따른 3D 프린팅방법은상기객체의 3D 메쉬데이터를입력받는단계, 상기 3D 메쉬데이터가상기 3D 프린터의출력공간에포함되는지판단하는단계, 상기 3D 프린터의출력공간에포함되지않는경우, 상기 3D 메쉬데이터를분할하여분할메쉬객체들을생성하는단계, 상기분할메쉬객체들을연결하기위한연결장치를형성하는단계, 상기분할메쉬객체들을상기출력공간에배치하는단계, 그리고배치된상기분할메쉬객체들을 3D 프린팅하는단계를포함한다.
Abstract translation: 一种用于3D打印物体的方法，该方法大于3D打印机的打印区域和使用该方法的装置。 3D打印方法包括接收对象的3D网格数据，确定3D网格数据是否包含在3D打印机的打印区域中，如果3D网格数据不包括在3D网格数据中，则通过分割3D网格数据来生成分割的网格对象 形成3D打印机的打印区域，形成用于连接分割网格对象的连接器，在打印区域中布置分割的网格对象，并以3D格式打印排列的分割网格物体。
-
公开(公告)号：KR1020140093836A
公开(公告)日：2014-07-29
申请号：KR1020130005703
申请日：2013-01-18
Applicant: 한국전자통신연구원
CPC classification number: G06K9/00221 , G06K9/00208 , G06K9/00275 , G06T15/00 , G06T17/00 , G06T2200/08 , G06T15/005
Abstract: Disclosed in the present invention are a method and an apparatus for creating a 3D montage. The apparatus for creating a 3D montage comprises: an image information extraction unit to extract image information from a face image to be reconstructed, using a face area based on statistical feature information and a feature vector; a 3D unique face reconstruction unit to reconstruct a 3D unique face model by fitting a 3D standard face model to face images of each view for the face image and feature information of each part for the face area; a 3D montage model creation unit to create a 3D montage model by combining the reconstructed 3D unique face model with 3D face expression model information and 3D decoration model information; and a montage image creation unit to create a montage image by projecting the created 3D montage model from each view.
Abstract translation: 在本发明中公开了一种用于创建3D蒙太奇的方法和装置。 用于创建3D蒙太奇的装置包括：图像信息提取单元，用于基于统计特征信息和特征向量，使用面部区域从要重建的面部图像提取图像信息; 3D独特面部重建单元，通过将3D标准面部模型拟合以面对面部图像的每个视图的图像和面部区域的每个部分的特征信息来重建3D独特面部模型; 3D蒙太奇模型创建单元，通过将重建的3D独特面部模型与3D面部表情模型信息和3D装饰模型信息相结合来创建3D蒙太奇模型; 以及蒙太奇图像创建单元，通过从每个视图投影创建的3D蒙太奇模型来创建蒙太奇图像。
-
公开(公告)号：KR1020140080723A
公开(公告)日：2014-07-01
申请号：KR1020120146287
申请日：2012-12-14
Applicant: 한국전자통신연구원
Abstract: Disclosed are a 3D avatar output device and a method thereof. The avatar output device is a vending machine shaped 3D avatar output device. The avatar output device comprises: an input information receiving part for receiving input information including at least one among user information, a 3D avatar theme, and 3D avatar output formation from a user; a restoration model producing part for producing a restoration model corresponding to a facial area in an image after acquiring the image of a user through a camera included in the 3D avatar output device; a genuine model producing part for producing a genuine model of a user based on the input information and restoration model; and an avatar output part for outputting a 3D avatar based on the 3D avatar output formation after producing the 3D avatar corresponding to the genuine model.
Abstract translation: 公开了一种3D头像输出装置及其方法。 化身输出装置是自动售货机形状的3D头像输出装置。 化身输出装置包括：输入信息接收部分，用于从用户接收包括用户信息，3D头像主题和3D头像输出形成中的至少一个的输入信息; 恢复模型生成部，用于通过包含在3D头像输出装置中的相机获取用户的图像，生成对应于图像中的面部区域的恢复模型; 基于输入信息和恢复模型来生产用户的真实模型的真实模型生成部分; 以及头像输出部件，用于在产生对应于真实模型的3D头像之后，基于3D头像输出形成输出3D头像。
-
公开(公告)号：KR1020140080703A
公开(公告)日：2014-07-01
申请号：KR1020120146013
申请日：2012-12-14
Applicant: 한국전자통신연구원
Abstract: Disclosed is a three-dimensional face restoration device. The three-dimensional face restoration device comprises an area extraction unit for extracting the face area of a user from an input image using a 3D template representing the face and depth information of the input image; and a shape restoration unit for restoring a three-dimensional face shape considering the accumulated depth values related to the face area of the user.
Abstract translation: 公开了一种立体面修复装置。 三维脸部恢复装置包括：区域提取单元，用于使用表示输入图像的脸部和深度信息的3D模板从输入图像中提取用户的脸部区域; 以及考虑到与用户的面部区域相关的累积深度值来恢复三维面部形状的形状恢复单元。
-
Patent Agency Ranking
公开(公告)号：KR1020130126234A
公开(公告)日：2013-11-20
申请号：KR1020120050168
申请日：2012-05-11
Applicant: 한국전자통신연구원
CPC classification number: G06T7/529 , G06T7/593 , G06T17/00 , G06T2200/08 , G06T2207/10012 , G06T2207/10024 , G06T2207/30201
Abstract: Disclosed are a device and method for recovering a 3-dimensional face based on multiple cameras. The device for recovering a 3-dimensional face based on multiple cameras according to the present invention comprises a multiple image analysis unit which inspects whether to synchronize with resolution of images inputted from multiple cameras, a texture image separation unit which separates an image for texture processing, an automatic synchronization unit of image for recovering which synchronizes images classified as asynchronous images, a 3-dimensional outer shape recovery unit which recovers to a 3-dimensional outer shape image by extracting depth information and by calculating 3-dimensional coordinate values and a texture processing unit which maps the image for texture processing on the 3-dimensional outer shape image. [Reference numerals] (200) Multiple images analyzing unit;(300) Texture image separation unit;(400) Automatic synchronizing unit for images to be restored;(500) 3-dimensional outer shape restoring unit;(600) Texture processing unit
Abstract translation: 公开了一种用于基于多个照相机来恢复三维面部的装置和方法。 根据本发明的用于恢复基于多个照相机的三维面部的装置包括：多图像分析单元，其检查是否与从多个照相机输入的图像的分辨率同步;纹理图像分离单元，其分离用于纹理处理的图像 ，用于恢复的图像的自动同步单元，其将分类为异步图像的图像同步;三维外形恢复单元，其通过提取深度信息恢复到三维外形图像，并且通过计算三维坐标值和纹理 处理单元，其将用于纹理处理的图像映射到三维外形图像上。 （200）多图像分析单元（300）纹理图像分离单元;（400）要恢复的图像的自动同步单元;（500）三维外形恢复单元;（600）纹理处理单元
-
公开(公告)号：KR101329057B1
公开(公告)日：2013-11-14
申请号：KR1020100027803
申请日：2010-03-29
Applicant: 한국전자통신연구원
CPC classification number: H04N21/2365 , H04N13/194 , H04N13/398 , H04N21/21805
Abstract: 본 발명은 다시점 입체 동영상 송신 장치 및 방법에 관한 것으로, 더욱 상세하게는 다양한 시점에서 촬영된 입체 영상 집합을 이용하여 다시점 입체 동영상을 생성하여 송신하는 다시점 입체 동영상 송신 장치 및 방법에 관한 것이다.
본 발명의 실시 예에 따른 다시점 입체 동영상 송신 방법은, 복수의 입체 영상 촬영기기로부터 촬영된 입체 영상 집합을 수신하는 단계; 상기 수신한 입체 영상 집합의 입체 프레임에서 하나 이상의 입체 프레임을 선택하고, 상기 선택된 입체 프레임을 순차적으로 배열하여 다시점 입체 동영상을 생성하는 단계; 상기 생성된 다시점 입체 동영상을 부호화하는 단계; 및 상기 부호화된 다시점 입체 동영상을 전송망을 통하여 송신하는 단계를 포함하되, 상기 다시점 입체 동영상을 생성하는 단계는, 시간 축과 공간 축 상에 입체 프레임으로 구성된 공간을 입체 프레임 단위로 이동시키는 단계를 포함한다.-
公开(公告)号：KR1020120067188A
公开(公告)日：2012-06-25
申请号：KR1020100128648
申请日：2010-12-15
Applicant: 한국전자통신연구원
Inventor： 김갑기
CPC classification number: G06T5/002 , G06T5/005 , G06T5/50 , G06T7/194 , G06T2207/10021 , G06T2207/20182 , G06T17/10 , G06T15/00 , G06T17/00 , G06T17/005 , G06T17/20 , G06T17/205 , G06T2200/04
Abstract: PURPOSE: A disparity map calibrating apparatus and a calibrating method thereof are provided to removing miscellaneous image generated in disparity map and filling hole due to blockade by using depth information of three dimensional models which is produced in previous frame. CONSTITUTION: A calibrating method of a disparity map calibrating apparatus comprises the following steps: dividing a domain into an inner part and outer part of an object by using three-dimensional model of binary image and previous frame by a disparity map domain setting unit(102); establishing the domain in disparity map of current frame by the disparity map domain setting unit; presuming pose of the object by a pose estimation unit(103); changing pose of the three-dimensional model of the previous frame by the pose estimation unit; obtaining depth information of two-dimensional projecting binary image of the three-dimensional model by the pose estimation unit ; and calibrating the disparity map by a disparity map compensation unit(104).
Abstract translation: 目的：提供视差图校准装置及其校准方法，以通过使用在前一帧中产生的三维模型的深度信息来去除由于封锁而在视差图和填充孔中产生的杂项图像。 构成：视差映射校准装置的校准方法包括以下步骤：通过使用视差映射域设置单元（102）的二进制图像和先前帧的三维模型将域划分为对象的内部和外部部分 ）; 通过视差映射域设置单位建立当前帧的视差图中的域; 通过姿势估计单元（103）来假设对象的姿势; 通过姿态估计单元改变前一帧的三维模型的姿态; 通过姿态估计单元获取三维模型的二维投影二值图像的深度信息; 以及通过视差图补偿单元（104）校准视差图。
-
公开(公告)号：KR1020110108551A
公开(公告)日：2011-10-06
申请号：KR1020100027803
申请日：2010-03-29
Applicant: 한국전자통신연구원
CPC classification number: H04N21/2365 , H04N13/194 , H04N13/398 , H04N21/21805
Abstract: 본 발명은 다시점 입체 동영상 송/수신 장치 및 방법에 관한 것으로, 보다 구체적으로, 다양한 시점에서 촬영된 입체 영상 집합을 이용하여 다시점 입체 동영상을 생성하여 송/수신하는 다시점 입체 동영상 송/수신 장치 및 방법에 관한 것이다.
본 발명의 실시 예에 따른 다시점 입체 동영상 송신 방법은, 복수의 입체 영상 촬영기기로부터 촬영된 입체 영상 집합을 수신하고, 상기 수신한 입체 영상 집합의 입체 프레임들 중 미리 결정된 방법에 따라 입체 프레임들을 선택하며, 상기 선택된 입체 프레임들을 순차적으로 배열하여 다시점 입체 동영상을 생성하고, 상기 생성된 다시점 입체 동영상을 부호화하여 전송망을 통하여 송신한다. 본 발명의 실시 예에 따른 다시점 입체 동영상 수신 방법은, 전송망을 통하여 다시점 입체 동영상을 수신하고, 상기 수신된 다시점 입체 동영상을 복호화하고, 상기 복호화된 다시점 입체 동영상을 디스플레이한다.-
公开(公告)号：KR100930715B1
公开(公告)日：2009-12-09
申请号：KR1020070107705
申请日：2007-10-25
Applicant: 한국전자통신연구원
IPC: G10L15/06 , G10L15/02 , G10L15/183 , G10L15/197
Abstract: A voice recognition method is provided to model various textual language phenomenons into statistical modeling among various knowledge sources. A morpheme is interpreted for a primitive text language corpus consisting of the separate words of Korean(S201). A morpheme language corpus separated is a separate word generated to morpheme. A word trigram which is the language model consisting of a morpheme unigram about a generated morpheme language corpus as described above, and bigram and trigrams is generated(S202). A first N - best recognition candidate to the maximum N is generated for a voice(S204). Recognition result candidates applying a morph-syntactic constraints are revaluated(S205). A second N-best list generated in above step is revaluated(S206). A final N-best list is generated.
Abstract translation: 提供语音识别方法来将各种文本语言现象模型化为各种知识源之间的统计建模。 一个语素被解释为由韩语的单词组成的原始文本语言语料库（S201）。 一个语素语言语料库是一个独立的单词，生成为语素。 生成如上所述的由关于所生成的语素语料库的词素单元构成的语言模型的词三元组，以及生成二元语法和三元语法（S202）。 为语音生成最大N的第一个N最佳识别候选者（S204）。 重新评估应用形态句法约束的识别结果候选（S205）。 在上述步骤中生成的第二个N-最佳列表被重新评估（S206）。 生成最终的N-最佳列表。
-
-
-
-
-
-
-
-
-
Search Results
64
records
(took 0.035 seconds)
