-
公开(公告)号：KR1020170030926A
公开(公告)日：2017-03-20
申请号：KR1020150128471
申请日：2015-09-10
Applicant: 한국과학기술원 , 코오롱글로벌 주식회사
IPC: H04N9/31
Abstract: 본발명은일반적인내부환경에서프로젝터와거울의정확한정렬없이간단한사용자입력만을이용하여캘리브레이션을수행할수 있도록하는단일프로젝터와구형거울기반의전방향프로젝션방법및 시스템에관한것으로, 상기방법은사용자에의해지정된환경기하정보를기반으로투사공간의기하좌표와프로젝터이미지의픽셀좌표간관계와상기투사공간의기하좌표와입력이미지픽셀좌표간관계를파악한후, 상기프로젝터이미지와상기입력이미지간픽셀좌표관계를정의하여캘리브레이션정보를생성하는전방향캘리브레이션단계; 및상기캘리브레이션정보를기반으로상기입력이미지를변형하여상기투사공간에전방향투사되는투사이미지를제작하는투사이미지제작단계를포함할수 있다.
Abstract translation: 本发明中，单个投影仪的方法和基于所述前向投影的方法和系统，其允许可以执行仅使用简单的用户输入而无需投影仪的精确对准，并在共同的内部环境中的反射镜的校准球面镜由用户给定的 环境，学习什么几何坐标和输入图像之间的关系，该像素的投影区域的几何坐标的坐标基于像素的几何信息坐标与投影图像的投影面积相互关系，限定像素坐标投影图像和输入图像之间的关系， 通过校准步骤产生校准信息的全向校准步骤; 并且基于校准信息修改输入图像以产生投影空间中的所有方向上投影的投影图像。
-
公开(公告)号：KR1020140122401A
公开(公告)日：2014-10-20
申请号：KR1020130039009
申请日：2013-04-10
Applicant: 한국과학기술원
IPC: G06T17/00
CPC classification number: G06T17/205 , G06K9/4609 , G06T2207/30201
Abstract: 3D 얼굴 영상 생성 방법 및 장치가 개시되어 있다. 3D(dimension) 얼굴 생성 방법은 2D 얼굴의 특징점을 기반으로 제1 3D 얼굴 모델을 제2 3D 얼굴 모델로 생성하는 단계와 제2 3D 얼굴 모델의 경계와 2D 얼굴의 경계를 매칭하는 단계를 포함할 수 있다. 2D 얼굴의 특징점은 2D 얼굴의 구조적인 특성에 대한 정보를 포함하는 점이고, 제2 3D 얼굴 모델은 제1 3D 얼굴 모델을 2D 얼굴의 구조적인 특성에 대한 정보를 반영하여 최종 피팅된 모델일 수 있다. 따라서, 2D 얼굴의 디테일한 부분에 대한 깊이 정보를 정확하게 예측하여 시각적인 왜곡이 적은 3D 얼굴을 생성할 수 있다.
Abstract translation: 公开了一种用于生成三维（3D）面部图像的方法和装置。 用于生成3D面部图像的方法可以包括基于2D面部的特征点产生具有第一3D面部模型的第二3D面部模型的步骤，以及匹配第二3D面部模型和边界的边界的步骤 的2D面。 2D面的特征点是包括关于2D面的结构特征的信息的点，并且第二3D面部模型可以是通过将关于2D面的结构特征的信息反映到第一3D面部模型的最终拟合模型 。 因此，可以通过精确地预测2D面的详细部分的深度信息来生成具有小的视觉失真的3D面。
-
公开(公告)号：KR1020140053721A
公开(公告)日：2014-05-08
申请号：KR1020120119987
申请日：2012-10-26
Applicant: 한국과학기술원
IPC: H04N13/00
CPC classification number: H04N13/128
Abstract: The present invention relates to an apparatus and a method for depth adjustment, capable of enabling a user to intuitively and easily adjust the depth of a 3D stereoscopic image. The apparatus comprises the following: a histogram analysis unit for acquiring a depth distribution map by analyzing a histogram of a disparity map corresponding to an inputted image; a Gaussian mixture model (GMM) fitting unit for acquiring a plurality of depth layers by the GMM fitting the depth distribution map; a disparity adjusting unit for adjusting at least one among location and volume of one of the depth layers and for outputting disparity mapping function reflecting the adjusted result; and a disparity map re-outputting unit for outputting a new disparity map after optimizing the disparity mapping function.
Abstract translation: 本发明涉及一种用于深度调节的装置和方法，能够使用户能够直观且容易地调整3D立体图像的深度。 该装置包括：直方图分析单元，用于通过分析对应于输入图像的视差图的直方图来获取深度分布图; 高斯混合模型（GMM）拟合单元，用于通过装配深度分布图的GMM获取多个深度层; 视差调整单元，用于调整深度层之一的位置和体积中的至少一个，并输出反映调整结果的视差映射函数; 以及视差图重新输出单元，用于在优化视差映射函数之后输出新的视差图。
-
公开(公告)号：KR101370785B1
公开(公告)日：2014-03-06
申请号：KR1020120125069
申请日：2012-11-06
Applicant: 한국과학기술원
IPC: H04N13/00
CPC classification number: G06T7/536 , H04N13/128 , H04N2013/0081
Abstract: The present invention relates to a depth map generating method of a stereoscopic image and an apparatus thereof capable of expressing the depth of an image in more detail by considering not only a vanishing point but also the detailed lines in the image. The method includes: a line segment grouping step to generate multiple line segments by grouping multiple edge pixels on an input image based on a density slope direction; a vanishing point detecting step to detect for at least one vanishing point in consideration of a merging result after merging the multiple line segments based on similarities; and a depth map generating step to generate an energy minimizing function which reflects a correlation between the line segments and the vanishing point and generating a depth map by decoding the energy minimizing function. [Reference numerals] (AA) Start; (BB) End; (S10) Line segment grouping; (S20) Vanishing point detection; (S30) Depth map generation
Abstract translation: 本发明涉及立体图像的深度图生成方法及其装置，其能够通过不仅考虑消失点而且还考虑图像中的详细线来更详细地表达图像的深度。 该方法包括：线段分组步骤，通过基于密度斜率方向对输入图像上的多个边缘像素进行分组来生成多个线段; 消失点检测步骤，基于相似度，在合并多个线段之后考虑到合并结果来检测至少一个消失点; 以及深度图生成步骤，用于生成反映线段与消失点之间的相关性的能量最小化函数，并通过解码能量最小化函数来生成深度图。 （附图标记）（AA）开始; （BB）结束; （S10）线段分组; （S20）消失点检测; （S30）深度图生成
-
Patent Agency Ranking
公开(公告)号：KR101370718B1
公开(公告)日：2014-03-06
申请号：KR1020120119988
申请日：2012-10-26
Applicant: 한국과학기술원
IPC: H04N13/00
CPC classification number: H04N13/128 , H04N5/23238 , H04N13/257 , H04N13/261
Abstract: The present invention includes: an image receiving unit which receives and stores an input image; a user interface which receives the input of a user who performs a 3D conversion operation; a panorama image generating unit which extracts feature points of multiple images which form an image sequence of the input image and generates a panorama image by warping and mixing the multiple images based on the extracted feature points; a depth setting unit which records a sign which includes depth information in at least one pixel among multiple pixels of the panorama image in response to a user input which is inputted through the user interface; a depth information transmitting unit which generates a panorama image depth map by calculating a depth value of all the pixels of the panorama image by calculating a depth value of the other pixels among the multiple pixels, based on the depth value of depth information of at least one pixel marked with the sign; a depth information remapping unit which generates an individual image depth map by mapping the depth value for each of the multiple images by using the depth map of the panorama image; and a stereoscopic image generating unit which generates a pair of stereoscopic images for each of the multiple images by using the individual image depth map and generates a stereoscopic image by using the generated pair of stereoscopic images. [Reference numerals] (AA) Start; (BB) End; (S110) Receive an input image; (S120) Generate a panorama image; (S130) Generate a reliability map; (S140) Receive a user sign; (S150) Assign depth information; (S160) Spread to the whole panorama image; (S170) Analyze color information; (S180) Generate a depth map; (S190) Remap an input image; (S200) Generate a pair of stereoscopic images; (S210) Generate a stereoscopic image
Abstract translation: 本发明包括：图像接收单元，其接收和存储输入图像; 用户接口，其接收执行3D转换操作的用户的输入; 全景图像生成单元，其提取形成所述输入图像的图像序列的多个图像的特征点，并且基于所提取的特征点通过扭曲和混合所述多个图像来生成全景图像; 深度设置单元，响应于通过用户界面输入的用户输入，在全景图像的多个像素中的至少一个像素中记录包括深度信息的符号; 深度信息发送单元，其通过基于深度信息的深度值至少计算多个像素中的其他像素的深度值来计算全景图像的所有像素的深度值来生成全景图像深度图 一个标有标志的像素; 深度信息重映射单元，其通过使用全景图像的深度图映射多个图像中的每一个的深度值来生成单独的图像深度图; 以及立体图像生成单元，其通过使用各个图像深度图生成所述多个图像中的每一个的一对立体图像，并且通过使用所生成的一对立体图像来生成立体图像。 （附图标记）（AA）开始; （BB）结束; （S110）接收输入图像; （S120）生成全景图像; （S130）生成可靠性图; （S140）接收用户标志; （S150）分配深度信息; （S160）传播到整个全景图像; （S170）分析颜色信息; （S180）生成深度图; （S190）重新输入图像; （S200）生成一对立体图像; （S210）生成立体图像
-
公开(公告)号：KR101356544B1
公开(公告)日：2014-02-19
申请号：KR1020120032207
申请日：2012-03-29
Applicant: 한국과학기술원
CPC classification number: G06T17/00 , G06T19/00 , H04N13/128
Abstract: 본 발명은 3차원 입체영상 생성 방법 및 장치에 관한 것으로, 상기 방법은 2D 평면영상에 2D 뎁스 맵 정보를 반영하여 적어도 하나의 3D 메쉬 서피스를 생성하는 단계; 상기 2D 평면영상에 3D 템플릿 모델을 반영하여 적어도 하나의 3D 입체 객체를 생성하는 단계; 상기 3D 메쉬 서피스와 3D 입체 객체를 3차원 공간상에 배치하고, 시점 고정시키는 단계; 상기 3D 메쉬 서피스와 3D 입체 객체의 입체감을 3차원 공간상에서 수정할 수 있도록 인터페이스를 제공하고, 상기 인터페이스를 통해 입력되는 제어값에 따라 상기 3D 메쉬 서피스와 3D 입체 객체의 입체감을 수정하는 단계; 및 상기 수정된 3D 메쉬 서피스와 3D 입체 객체를 적어도 두개의 카메라를 통해 촬영하여 3D 입체 영상을 획득하는 단계를 포함할 수 있다.
Abstract translation: 提供了一种用于生成3D立体图像的方法，其包括：通过将2D深度图信息应用于2D平面图像来生成至少一个3D网格表面; 通过将3D模板模型应用于2D平面图像来生成至少一个3D固体对象; 将3D网格表面和3D实体对象布置在3D空间上并固定视点; 提供界面，使得3D网格表面和3D固体对象的三次效果在3D空间上是可校正的，并且根据通过界面输入的控制值来校正3D网格表面和3D固体对象的立方效果; 以及通过用至少两个照相机拍摄校正的3D网格表面和3D固体物体来获得3D立体图像。
-
公开(公告)号：KR1020130110339A
公开(公告)日：2013-10-10
申请号：KR1020120032207
申请日：2012-03-29
Applicant: 한국과학기술원
CPC classification number: G06T17/00 , G06T19/00 , H04N13/128
Abstract: PURPOSE: A method and a device for generating a three dimensional stereoscopic image are provided to improve work efficiency by performing a depth information correction method of using a template form and a depth information correction method of a user definition. CONSTITUTION: A method for generating a three dimensional (3D) stereoscopic image comprises the following steps: a step (S13) aims to produce at least one 3D mesh surface by reflecting two dimensional (2D) depth map information to a 2D monoscopic image; a step (S14) aims to produce at least one 3D cubic object by reflecting a 3D template model to the 2D monoscopic image; a step (S15) aims to arrange the 3D mesh surface and the 3D cubic object on a 3D space image, and to fix a viewpoint; a step (S16) aims to provide an interface in order to modify a stereoscopic perception of the 3D mesh surface and the 3D cubic object on the 3D space, and to modify the stereoscopic perception of the 3D mesh surface and the 3D cubic object according to a control value inputted through the interface; and a step (S18) aims to acquire the 3D stereoscopic image by taking a picture of the modified 3D mesh surface and the modified 3D cubic object through at least two cameras.
Abstract translation: 目的：提供一种用于生成三维立体图像的方法和装置，以通过执行使用用户定义的模板形式和深度信息校正方法的深度信息校正方法来提高工作效率。 构成：用于生成三维（3D）立体图像的方法包括以下步骤：步骤（S13）旨在通过将二维（2D）深度图信息反射到2D单视图图像来产生至少一个3D网格表面; 步骤（S14）旨在通过将3D模板模型反射到2D单视场图像来产生至少一个3D立方体对象; 步骤（S15）旨在将3D网格表面和3D立方体对象布置在3D空间图像上，并且固定视点; 步骤（S16）旨在提供一种接口，以便修改3D空间上的3D网格表面和3D立方体对象的立体感觉，并根据以下方式修改3D网格表面和3D立方体对象的立体感知 通过界面输入的控制值; 并且步骤（S18）旨在通过至少两个照相机拍摄修改的3D网格表面和修改的3D立方体对象的图片来获取3D立体图像。
-
公开(公告)号：KR101208767B1
公开(公告)日：2012-12-05
申请号：KR1020110037956
申请日：2011-04-22
Applicant: 한국과학기술원
IPC: H04N13/00
Abstract: 곡면투사를이용한입체영상생성방법, 장치및 시스템, 이를위한기록매체를공개한다. 본발명은두 개이상의가상카메라시점에서관측되는모든영역을하나의가상카메라시점으로관측할수 있도록하는곡면투사면을생성하고, 곡면투사면을통해이미지와깊이정보에대해한 번의굴절렌더링을수행하여폐색에의한정보손실없이입체영상생성을위한시간을크게줄일수 있으며, 입체영상의양안시차를용이하게조절할수 있다.
-
公开(公告)号：KR1020090115315A
公开(公告)日：2009-11-05
申请号：KR1020080041098
申请日：2008-05-02
Applicant: 한국과학기술원
IPC: G02F1/1335 , G02F1/167
CPC classification number: G02F1/133555 , G02B27/26 , G09G2320/0626
Abstract: PURPOSE: A display apparatus is provided to increase the contrast ratio of an image by selectively adjusting the permeability of a desired area in a transparent image panel. CONSTITUTION: According to an electric signal, a transparent image panel(200) displays an image. Blind panels(210) control the transmittance so that the light transmitted through the transparent image panel can be selectively transmitted by being formed on the front side and back side of the transparent image panel. The blind panels control the transmittance of the light transmitted through one of the front side and back side of the transparent image panel, generates an image corresponding to the image displayed on the transparent image panel, and adjusts the transmittance of the light transmitted through the generated image.
Abstract translation: 目的：提供一种显示装置，通过选择性地调节透明图像面板中的期望区域的透过率来增加图像的对比度。 构成：根据电信号，透明图像面板（200）显示图像。 遮光板（210）控制透射率，透过透明图像面板的光可以通过形成在透明图像面板的前侧和后侧而选择性地透射。 盲板控制透过透明图像面板的正面和背面中的一个透射的透射率，产生与透明图像面板上显示的图像相对应的图像，并且调节通过产生的透射图像透射的透射率 图片。
-
-
-
-
-
-
-
-
Search Results
29
records
(took 0.03 seconds)
