-
公开(公告)号：KR100924121B1
公开(公告)日：2009-10-29
申请号：KR1020070131827
申请日：2007-12-15
Applicant: 한국전자통신연구원
IPC: H04N9/64
CPC classification number: H04N17/02 , G06T7/80 , H04N13/133 , H04N13/239 , H04N13/243 , H04N13/246 , H04N13/25 , H04N13/257 , H04N17/002
Abstract: 본 발명은 컬러 체크 챠트를 이용한 다중 카메라간 컬러 일치 방법에 관한 것이다. 즉, 본 발명에서는 3차원 정보 복원을 수행하는 영상처리 기법에서 여러 대의 카메라를 사용하여 촬영한 영상을 조합하는 경우 카메라의 특성 또는 촬영 시 조명의 조건에 따라 발생하는 색 차에 대해 각 카메라 별로 기준 컬러 모델과의 관계를 모델링하여 보정을 수행함으로써 다중 카메라간 컬러를 일치시킨다. 이에 따라 여러 대의 카메라를 사용하여 획득한 영상들을 바탕으로 3차원 정보를 복원하는 경우, 카메라마다 다른 컬러 값들을 동일한 값으로 보정하여 2차원 영상들의 정보를 효과적으로 조합하는 것이 가능하다. 또한, 한 대의 카메라를 옮겨가며 촬영하는 것이 불가능한 상황인 동적 물체의 3차원 정보 복원에 유용하다.
카메라, 컬러, 일치, 컬러 체커 챠트Abstract translation: 使用颜色检查图的多视点相机颜色校准方法包括测量从多个相机输入的照片数据的亮度范围; 拍摄每个相机的颜色检查图，并将图表的亮度值调整在每个相机的输入图像的亮度范围内; 通过使用格子图的灰色补丁校正每个相机的亮度; 对由每个相机获得的检查图的颜色值与由颜色检查图提供的标准颜色值之间的关系进行建模; 并且通过建模来校正颜色差异，以校准多个相机之间的颜色。 因此，能够恢复不能通过仅移动一个照相机而拍摄的摄影用移动物体的三维信息。
-
公开(公告)号：KR1020090065357A
公开(公告)日：2009-06-22
申请号：KR1020070132859
申请日：2007-12-17
Applicant: 한국전자통신연구원
IPC: G06T17/00
CPC classification number: G06T17/10 , G06F17/5009 , G06T2210/21 , G06T2210/56
Abstract: A collision resolution modeling method between a particle and an object in a particle system is provided to flexibly control forces acting on a collision resolution method according to attributes of the particle system. With regard to a particle and a boundary particle, a bounce force acting in inverse proportion to a distance between the two objects is calculated. A repulsive force for modeling a non-bounce collision is calculated by changing a normal component of a force and speed acting on the particle. A damping force for the boundary particle and the particle is calculated. The calculated bounce force, repulsive force, and damping force are controlled to calculate overall forces for a particle system(100).
Abstract translation: 提供粒子系统中的粒子和物体之间的碰撞解析建模方法，以根据粒子系统的属性灵活地控制作用于碰撞解决方法的力。 对于粒子和边界粒子，计算与两个物体之间的距离成反比的反弹力。 通过改变作用在粒子上的力和速度的正常分量来计算用于建模非反弹碰撞的排斥力。 计算边界粒子和粒子的阻尼力。 控制计算的反弹力，排斥力和阻尼力以计算粒子系统（100）的总体力。
-
公开(公告)号：KR1020090065354A
公开(公告)日：2009-06-22
申请号：KR1020070132855
申请日：2007-12-17
Applicant: 한국전자통신연구원
CPC classification number: G06T11/60
Abstract: An image conversion method for artistic effect application of a 2D image and a device therefore are provided to perform image conversion through various artistic techniques by image conversion processing using a computer, thereby offering a natural image. A rendering server(101) converts a 2D image requested to be changed by a user into a required type of image through non-realistic rendering. When an image change request is received, a web server(102) delivers the image change request to the rendering server according to the user's terminal specification. The web server delivers a result image to the corresponding user's terminal. The web server delivers an image change program to the user's terminal. The web server controls so that an image change function is executed in the corresponding user's terminal. The user's terminal demands non-realistic rendering processing for the 2D image to the web server.
Abstract translation: 因此，提供了一种用于2D图像和装置的艺术效果应用的图像转换方法，通过使用计算机的图像转换处理通过各种艺术技术执行图像转换，从而提供自然图像。 呈现服务器（101）通过非现实呈现将被请求改变的2D图像转换成所需类型的图像。 当接收到图像改变请求时，web服务器（102）根据用户的终端规范将图像改变请求传递给呈现服务器。 网络服务器将结果图像传递给相应的用户终端。 网络服务器向用户终端传送图像更改程序。 Web服务器进行控制，使得在对应的用户终端中执行图像改变功能。 用户的终端需要2D图像到Web服务器的不切实际的渲染处理。
-
公开(公告)号：KR1020090065101A
公开(公告)日：2009-06-22
申请号：KR1020070132545
申请日：2007-12-17
Applicant: 한국전자통신연구원
Abstract: An accuracy measurement method of a 3D graphic model by using an image and a device thereof are provided to compare a reference image that takes a picture of a real thing with a 3D model which is being made, thereby visually notifying a designer of a part from which accuracy and error are created. A camera corrector(102) generates camera parameters by calculating a location, a direction, and an internal factor of a camera, from a reference image. A model synthesizer(104) calculates a location and a direction of a 3D model based on a matching relation between the reference image and 3D graphic model data. A rendering unit(106) performs a rendering process of the 3D graphic model by using the camera parameters, and generates a synthesis image. An image feature extractor(108) extracts features of the synthesis image and the reference image. An error calculator(110) calculates distance and length errors.
Abstract translation: 提供了通过使用图像及其装置的3D图形模型的精度测量方法来比较将实物的图片与正在制作的3D模型进行比较的视觉通知设计者的参考图像， 创建了哪些精度和错误。 相机校正器（102）通过从参考图像计算相机的位置，方向和内部因素来生成相机参数。 模型合成器（104）基于参考图像和3D图形模型数据之间的匹配关系来计算3D模型的位置和方向。 渲染单元（106）通过使用相机参数来执行3D图形模型的渲染处理，并且生成合成图像。 图像特征提取器（108）提取合成图像和参考图像的特征。 误差计算器（110）计算距离和长度误差。
-
公开(公告)号：KR100901276B1
公开(公告)日：2009-06-09
申请号：KR1020070132678
申请日：2007-12-17
Applicant: 한국전자통신연구원
IPC: G06T15/10
CPC classification number: G06T15/02
Abstract: An effect processing unit using a style line capable of expressing the effect that a person seems to draw is provided to determine the shape of the directly in-line by three control points and generate a curve by reflecting the curvature of both ends and the curve based on three control points. A contour line generating unit(100) extracts an edge list in a polygon of a 3D object. The contour line generating unit produces the contour line by using the location information of the camera and the extracted edge list. The style line generating unit(110) groups the extracted edge list. The style line generating unit sets up each group as one line. The style line generating unit produces each by group style line by transforming the set up each line. An effect processing unit(120) expresses the line style by inserting the style line into inside and outside of the contour line.
Abstract translation: 提供一种使用能够表达人似乎绘制的效果的样式线的效果处理单元，以通过三个控制点直接确定直线的形状，并且通过反映两端的曲率和曲线来生成曲线 在三个控制点。 轮廓线生成单元（100）提取3D对象的多边形中的边缘列表。 轮廓线生成单元通过使用摄像机的位置信息和提取的边缘列表来生成轮廓线。 样式线生成单元（110）对所提取的边缘列表进行分组。 样式线生成单元将每组设置为一行。 样式线生成单元通过组合样式线生成每一行，通过转换每行的设置。 效果处理单元（120）通过将样式线插入轮廓线的内部和外部来表示线条样式。
-
Patent Agency Ranking
公开(公告)号：KR1020090050910A
公开(公告)日：2009-05-20
申请号：KR1020070132854
申请日：2007-12-17
Applicant: 한국전자통신연구원
Abstract: 본 발명은 디지털 만화책 제작 방법 및 장치에 관한 것이다. 즉, 본 발명에서는 영화와 같은 동영상이나 주변에서 쉽게 구할 수 있는 사진 등을 입력으로 하여 영상들을 만화 영상과 같이 렌더링하는 카투닝(cartooning)과 만화책처럼 보이게 하는 양식화(stylization) 작업, 그리고 말풍선 등 만화적 요소를 배치하는 작업을 통해 사용자의 수작업을 최소로 하여 누구나 쉽게 디지털 만화책을 제작할 수 있도록 한다.
만화, 디지털, 랜더링, 카투닝-
公开(公告)号：KR1020080052272A
公开(公告)日：2008-06-11
申请号：KR1020070071177
申请日：2007-07-16
Applicant: 한국전자통신연구원
Abstract: A cartoon animation production method using a character animation and mesh modification and a system therefor are provided to produce an animation which is more similar to a 2D cartoon character and is full of movement by using real 3D character animation data. A motion analyzing module(100) receives existing motion data(20) which are data of motion which is not modified in a character, analyzes an animation value having respective joints of a character from the existing motion data, and extracts a parameter. A mesh modification module(200) receives existing mesh data indicating an external shape of the character, the parameter, and existing skinning data which are combined information so that a bone closely adheres to a mesh, and generates modified mesh data(40). A motion modification module(300) receives the existing motion data and modifies motion by using the parameter. A skinning module(400) receives the modified mesh data, the modified motion data, and the existing skinning data, and finally generates character animation data having the motion of a cartoon style.
Abstract translation: 提供了使用角色动画和网格修改及其系统的卡通动画制作方法，以产生与2D卡通人物更相似的动画，并且通过使用真实的3D角色动画数据充满了移动。 运动分析模块（100）接收作为未被修改的运动数据的现有运动数据（20），从现有运动数据分析具有各个角色的关节的动画值，并提取参数。 网格修改模块（200）接收指示字符的外部形状的现有网格数据，所述参数和现有的皮肤数据是组合信息，使得骨骼紧密地粘附到网格上，并且生成修改的网格数据（40）。 运动修改模块（300）通过使用参数接收现有运动数据并修改运动。 剥皮模块（400）接收修改的网格数据，修改的运动数据和现有的皮肤数据，并且最终生成具有漫画风格的运动的角色动画数据。
-
公开(公告)号：KR1020080051026A
公开(公告)日：2008-06-10
申请号：KR1020070100389
申请日：2007-10-05
Applicant: 한국전자통신연구원
IPC: H04N5/232
Abstract: An apparatus and a method for estimating a focus length of a camera are provided to calculate the fixed focus length of a zoom lens camera from the positions of the corresponding points between images and to verify the reliability of the calculated focus length easily. An image corresponding point position designation member(110) designates the corresponding point positions between the inputted two dimensional image pair. A cost function defining member(120) defines a cost function in order to calculate the focus length for performing the automatic correction of a camera from the corresponding point positions between the designated image pair. A cost function minimizing member(130) calculates the optimum focus length for minimizing the defined cost function. A position and posture calculating member calculates the photographing position and posture of a camera through an algorithm including 5-points, 6-points, 7-points and 8-points by means of the focus length and the corresponding point positions between image pair.
Abstract translation: 提供用于估计相机的焦距的装置和方法，以从图像之间的对应点的位置计算变焦镜头相机的固定焦距，并且容易地验证所计算的焦距长度的可靠性。 图像对应点位置指定构件（110）指定输入的二维图像对之间的对应点位置。 成本函数定义构件（120）定义成本函数，以便从指定图像对之间的相应点位置计算用于执行相机自动校正的焦距。 最小化成员（130）的成本函数计算用于使定义的成本函数最小化的最佳聚焦长度。 位置和姿势计算部件通过包括5点，6点，7点和8点的算法，通过焦点长度和图像对之间的对应点位置来计算照相机的拍摄位置和姿势。
-
公开(公告)号：KR1020080050970A
公开(公告)日：2008-06-10
申请号：KR1020070074065
申请日：2007-07-24
Applicant: 한국전자통신연구원
IPC: G06T13/60
CPC classification number: G06T13/60 , G06T2210/24 , G06T2210/56
Abstract: An apparatus and a method for expressing wave and bubble using water particle data are provided to create an image of various water scenes, major factor of virtual special effects, and help produce a detailed and rich expression of animation. A water particle data analyzing module(101) agglomerates water particles by certain distance by using position information of inputted water particle, classifies main water clods and small water beads, and analyzes distribution of the agglomerated water particles. A wave generating module(102) generates wave particles at a position where the main water clods and the small water beads are gathered by using the agglomerated water particles. A bubble generating module(103) obtains the curvature of the water surface where the water particles are positioned, and generates bubble particles by using a change in the speed of the water particles and the curvature. A wave-bubble movement and extinction module(104) obtains a flow variation of the water particles adjacent to the generated water and bubble particles, and moves the generated wave and bubble particles by the obtained flow variation.
Abstract translation: 提供了一种使用水粒子数据表达波浪和气泡的装置和方法，以创建各种水景的图像，虚拟特效的主要因素，并有助于产生动画的详尽丰富的表现。 水粒子数据分析模块（101）通过使用输入水粒子的位置信息将水粒子聚集一定距离，对主水域和小水珠进行分类，分析凝聚水粒子的分布。 波发生模块（102）通过使用附聚水粒子在主水槽和小水珠聚集的位置处产生波粒子。 气泡生成模块（103）获得水粒子的水面的曲率，通过使用水粒子的速度和曲率的变化来生成气泡粒子。 波泡运动和消光模块（104）获得与产生的水和气泡颗粒相邻的水颗粒的流动变化，并且通过获得的流动变化来移动产生的波和气泡颗粒。
-
公开(公告)号：KR1020080041978A
公开(公告)日：2008-05-14
申请号：KR1020070060583
申请日：2007-06-20
Applicant: 한국전자통신연구원 , 중앙대학교 산학협력단
CPC classification number: G06T5/005 , G06T11/00 , G06T11/001
Abstract: A painterly rendering method based on a painter's painting process and an exhibition system thereof are provided to understand the drawing manner by a human being and to reproduce the unavoidable errors by the drawing process. A camera(104) obtains a photograph of a visitor. A computer(101) performs a rendering process of the visitor's photograph obtained by the camera. A display(103) is attached to an exhibition system and displays the rendering process. A printer(106) outputs the result changed by the computer as the painterly image by means of the obtained photograph. The computer divides the error generation points occurred by the painting process into three stages such as a point due to the mistake of eyes, a point due to the mistake of hands and a point due to the mistake of discrimination, then performs the rendering process and changes the input image into the painterly image.
Abstract translation: 提供了一种基于画家绘画过程的绘画方法及其展示系统，以了解人的绘画方式，并通过绘画过程再现不可避免的错误。 照相机（104）获取访客的照片。 计算机（101）执行摄像机获取的访客照片的呈现处理。 显示器（103）附接到展示系统并显示渲染过程。 打印机（106）通过所获得的照片将由计算机改变的结果作为画家图像输出。 计算机将绘画过程发生的错误生成点分为三个阶段，如由于眼睛的错误导致的一个点，由于手的错误以及由于歧视的错误导致的点，则执行渲染过程和 将输入图像更改为画面图像。
-
-
-
-
-
-
-
-
-
Search Results
274
records
(took 0.025 seconds)
